id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 6
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-06-27 00:00:00
| authors_parsed
sequencelengths 1
2.83k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.09464 | Ruby Kumari | Ruby Kumari, Gaurav Purohit and Abhijit Karmakar | Efficient Hardware Implementation of Modular Multiplier over GF (2m) on FPGA | 18 pages, 11 figures | null | null | null | cs.CR cs.AR | http://creativecommons.org/licenses/by/4.0/ | Elliptic curve cryptography (ECC) has emerged as the dominant public-key protocol, with NIST standardizing parameters for binary field GF(2^m) ECC systems. This work presents a hardware implementation of a Hybrid Multiplication technique for modular multiplication over binary field GF(2m), targeting NIST B-163, 233, 283, and 571 parameters. The design optimizes the combination of conventional multiplication (CM) and Karatsuba multiplication (KM) to enhance elliptic curve point multiplication (ECPM). The key innovation uses CM for smaller operands (up to 41 bits for m=163) and KM for larger ones, reducing computational complexity and enhancing efficiency. The design is evaluated in three areas: Resource Utilization For m=163, the hybrid design uses 6,812 LUTs, a 39.82% reduction compared to conventional methods. For m=233, LUT usage reduces by 45.53% and 70.70% compared to overlap-free and bit-parallel implementations. Delay Performance For m=163, achieves 13.31ns delay, improving by 37.60% over bit-parallel implementations. For m=233, maintains 13.39ns delay. Area-Delay Product For m=163, achieves ADP of 90,860, outperforming bit-parallel (75,337) and digit-serial (43,179) implementations. For m=233, demonstrates 16.86% improvement over overlap-free and 96.10% over bit-parallel designs. Results show the hybrid technique significantly improves speed, hardware efficiency, and resource utilization for ECC cryptographic systems. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 07:14:05 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 07:23:12 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Jun 2025 05:32:36 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Kumari",
"Ruby",
""
],
[
"Purohit",
"Gaurav",
""
],
[
"Karmakar",
"Abhijit",
""
]
] |
2506.09679 | Andrew Gracyk | Andrew Gracyk | Geometric flow regularization in latent spaces for smooth dynamics with the efficient variations of curvature | Small improvements to some mathematical arguments | null | null | null | math.NA cs.NA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We design strategies in nonlinear geometric analysis to temper the effects of adversarial learning for sufficiently smooth data of numerical method-type dynamics in encoder-decoder methods, variational and deterministic, through the use of geometric flow regularization. We augment latent spaces with geometric flows to control structure. Our techniques rely on adaptations of curvature and Ricci flow. We invent new geometric flows or discover them neurally and non-parametrically. All of our flows are solved using physics-informed learning. Traditional geometric meaning is traded for computing ability, but we maintain key geometric invariants, the primary of which are maintained, intrinsically-low structure, canonicity or a lack of irregularity, nontriviality due to sufficient lower bounds on curvature, and distortion of volume element, that develop quality in the inference stage. Our primary contributions are fourfold. We develop a loss based on Gaussian curvature using closed path circulation integration for surfaces, bypassing automatic differentiation of the Christoffel symbols through use of Stokes' theorem. We invent a new parametric flow derived from a linear version of the Gauss equation and a Riemannian decomposition for a custom tensor defined with a normal Hessian and Weyl tensor proxies. We develop two strategies based on time differentiation of functionals, one with a special case of scalar curvature for conformally-changed metrics, and another with harmonic maps, their energy, and induced metrics. Our methods, while diminished analytically, maintain overall integral latent structure. We showcase that curvature flows and the formulation of geometric structure in intermediary encoded settings enhance learning and overall zero-shot and adversarial fidelity. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 12:53:09 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 11:16:20 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Gracyk",
"Andrew",
""
]
] |
2506.09725 | Hiroyuki Tajima | Tingyu Zhang, Hiroyuki Tajima, Motoi Tachibana | Schwinger-Keldysh approach to tunneling transport at a hadron-quark interface | 9 pages, 4 figures | null | null | RIKEN-iTHEMS-Report-25 | nucl-th astro-ph.HE cond-mat.supr-con hep-ph hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We theoretically discuss quantum tunneling transport and frictions at a hadron-quark matter interface based on the Schwinger-Keldysh approach combined with the tunneling Hamiltonian, which has been developed in the context of condensed matter physics. In the inner core of massive neutron stars, it is expected that cold quark matter appears at sufficiently high densities and hence exhibits color superconductivity, surrounded by nucleon superfluids at lower densities. The perturbative expressions of the tunneling current and the friction at the interface are obtained in terms of the non-equilibrium Green's functions. We demonstrate the DC Josephson current that occurs at the hadron-quark superfluid interface in the present scheme. Our framework can be applied to various conflagrations involving the interfaces relevant to astrophysical phenomena. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 13:34:05 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 07:15:36 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Zhang",
"Tingyu",
""
],
[
"Tajima",
"Hiroyuki",
""
],
[
"Tachibana",
"Motoi",
""
]
] |
2506.10046 | Gamal G.L. Nashed | G. G. L. Nashed and Emmanuel N. Saridakis | 3-dimensional charged black holes in $f({Q})$ gravity | 12 pages, four figures, updating references | null | null | null | gr-qc hep-th | http://creativecommons.org/licenses/by/4.0/ | We present new exact charged black hole solutions in (2+1) dimensions within the framework of $f({Q})$ gravity, where ${Q}$ denotes the non-metricity scalar. By considering a cubic $f({Q})$ form we derive classes of charged and uncharged spherically symmetric solutions, and we identify conditions under which these reduce to the well-known Banados-Teitelboim-Zanelli (BTZ) black hole. Notably, our analysis reveals a novel charged solution that is asymptotically Anti-de Sitter (AdS) but cannot be continuously deformed into a GR counterpart, and thus arising purely from the
higher-order nature of the non-metricity corrections. Furthermore, we explore the geometrical properties of the solutions, demonstrating the existence of multiple horizons and a softer central singularity compared to GR. Additionally, we calculate various thermodynamic quantities, such as Hawking temperature, entropy, and heat capacity, with results confirming thermal stability. Finally, a detailed study of the geodesic motion and the effective potentials unveils stable photon orbits, as well as the effect of cubic non-metricity corrections on orbital dynamics. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 08:00:24 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 05:27:16 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Nashed",
"G. G. L.",
""
],
[
"Saridakis",
"Emmanuel N.",
""
]
] |
2506.10049 | Francesco Vinci | Francesco Vinci, Gyunam Park, Wil van der Aalst, Massimiliano de Leoni | Online Discovery of Simulation Models for Evolving Business Processes (Extended Version) | null | null | null | null | cs.SE cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Business Process Simulation (BPS) refers to techniques designed to replicate the dynamic behavior of a business process. Many approaches have been proposed to automatically discover simulation models from historical event logs, reducing the cost and time to manually design them. However, in dynamic business environments, organizations continuously refine their processes to enhance efficiency, reduce costs, and improve customer satisfaction. Existing techniques to process simulation discovery lack adaptability to real-time operational changes. In this paper, we propose a streaming process simulation discovery technique that integrates Incremental Process Discovery with Online Machine Learning methods. This technique prioritizes recent data while preserving historical information, ensuring adaptation to evolving process dynamics. Experiments conducted on four different event logs demonstrate the importance in simulation of giving more weight to recent data while retaining historical knowledge. Our technique not only produces more stable simulations but also exhibits robustness in handling concept drift, as highlighted in one of the use cases. | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 14:57:05 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 08:14:12 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Vinci",
"Francesco",
""
],
[
"Park",
"Gyunam",
""
],
[
"van der Aalst",
"Wil",
""
],
[
"de Leoni",
"Massimiliano",
""
]
] |
2506.10082 | Chenjian Gao | Chenjian Gao, Lihe Ding, Xin Cai, Zhanpeng Huang, Zibin Wang, Tianfan Xue | LoRA-Edit: Controllable First-Frame-Guided Video Editing via Mask-Aware LoRA Fine-Tuning | 12 pages | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Video editing using diffusion models has achieved remarkable results in generating high-quality edits for videos. However, current methods often rely on large-scale pretraining, limiting flexibility for specific edits. First-frame-guided editing provides control over the first frame, but lacks flexibility over subsequent frames. To address this, we propose a mask-based LoRA (Low-Rank Adaptation) tuning method that adapts pretrained Image-to-Video (I2V) models for flexible video editing. Our approach preserves background regions while enabling controllable edits propagation. This solution offers efficient and adaptable video editing without altering the model architecture. To better steer this process, we incorporate additional references, such as alternate viewpoints or representative scene states, which serve as visual anchors for how content should unfold. We address the control challenge using a mask-driven LoRA tuning strategy that adapts a pre-trained image-to-video model to the editing context. The model must learn from two distinct sources: the input video provides spatial structure and motion cues, while reference images offer appearance guidance. A spatial mask enables region-specific learning by dynamically modulating what the model attends to, ensuring that each area draws from the appropriate source. Experimental results show our method achieves superior video editing performance compared to state-of-the-art methods. Project Page: https://cjeen.github.io/LoraEditPaper | [
{
"version": "v1",
"created": "Wed, 11 Jun 2025 18:03:55 GMT"
},
{
"version": "v2",
"created": "Fri, 20 Jun 2025 02:52:21 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Jun 2025 15:22:08 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Gao",
"Chenjian",
""
],
[
"Ding",
"Lihe",
""
],
[
"Cai",
"Xin",
""
],
[
"Huang",
"Zhanpeng",
""
],
[
"Wang",
"Zibin",
""
],
[
"Xue",
"Tianfan",
""
]
] |
2506.10412 | Ching Chang | Ching Chang, Jeehyun Hwang, Yidan Shi, Haixin Wang, Wen-Chih Peng, Tien-Fu Chen, Wei Wang | Time-IMM: A Dataset and Benchmark for Irregular Multimodal Multivariate Time Series | This paper is currently under review | null | null | null | cs.LG cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | Time series data in real-world applications such as healthcare, climate modeling, and finance are often irregular, multimodal, and messy, with varying sampling rates, asynchronous modalities, and pervasive missingness. However, existing benchmarks typically assume clean, regularly sampled, unimodal data, creating a significant gap between research and real-world deployment. We introduce Time-IMM, a dataset specifically designed to capture cause-driven irregularity in multimodal multivariate time series. Time-IMM represents nine distinct types of time series irregularity, categorized into trigger-based, constraint-based, and artifact-based mechanisms. Complementing the dataset, we introduce IMM-TSF, a benchmark library for forecasting on irregular multimodal time series, enabling asynchronous integration and realistic evaluation. IMM-TSF includes specialized fusion modules, including a timestamp-to-text fusion module and a multimodality fusion module, which support both recency-aware averaging and attention-based integration strategies. Empirical results demonstrate that explicitly modeling multimodality on irregular time series data leads to substantial gains in forecasting performance. Time-IMM and IMM-TSF provide a foundation for advancing time series analysis under real-world conditions. The dataset is publicly available at https://www.kaggle.com/datasets/blacksnail789521/time-imm/data, and the benchmark library can be accessed at https://anonymous.4open.science/r/IMMTSF_NeurIPS2025. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 07:07:22 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 21:10:15 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Chang",
"Ching",
""
],
[
"Hwang",
"Jeehyun",
""
],
[
"Shi",
"Yidan",
""
],
[
"Wang",
"Haixin",
""
],
[
"Peng",
"Wen-Chih",
""
],
[
"Chen",
"Tien-Fu",
""
],
[
"Wang",
"Wei",
""
]
] |
2506.10451 | David Kubiznak | Ivan Kol\'a\v{r}, David Kubiznak, Poula Tadros | Rotating Carroll Black Holes: A No Go Theorem | 11 pages, no figures v2: improved 3d discussion, added references | null | null | null | hep-th gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recently, there has been a lot of interest in Carroll black holes and in particular whether or not one could find a Carrollian analogue of a rotating black hole spacetime. Here we show that every stationary and axisymmetric solution (and thence also a black hole) of Carrollian general relativity in any number of $d>3$ dimensions is necessarily also static (up to a "topological rotation"). The case of $d=3$ dimensions is special. There, the topological rotation is important and one can have a rotating Carroll BTZ black hole, obtained from a static one by the Carroll boost accompanied by the re-identification of the angular coordinate, similar to what happens in the Lorentzian case. We also find a Carrollian analogue of an accelerating black hole, showing that Schwarzschild is not the only possible stationary and axisymmetric Carroll black hole in four dimensions. A generalization of the no go theorem to include Maxwell, dilatonic, and axionic matter fields is also discussed. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 07:57:06 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 11:02:13 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Kolář",
"Ivan",
""
],
[
"Kubiznak",
"David",
""
],
[
"Tadros",
"Poula",
""
]
] |
2506.10872 | Ziv Scully | Ziv Scully, Alexander Terenin | The Gittins Index: A Design Principle for Decision-Making Under Uncertainty | null | null | null | null | math.OC cs.LG cs.PF math.PR stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Gittins index is a tool that optimally solves a variety of decision-making problems involving uncertainty, including multi-armed bandit problems, minimizing mean latency in queues, and search problems like the Pandora's box model. However, despite the above examples and later extensions thereof, the space of problems that the Gittins index can solve perfectly optimally is limited, and its definition is rather subtle compared to those of other multi-armed bandit algorithms. As a result, the Gittins index is often regarded as being primarily a concept of theoretical importance, rather than a practical tool for solving decision-making problems.
The aim of this tutorial is to demonstrate that the Gittins index can be fruitfully applied to practical problems. We start by giving an example-driven introduction to the Gittins index, then walk through several examples of problems it solves - some optimally, some suboptimally but still with excellent performance. Two practical highlights in the latter category are applying the Gittins index to Bayesian optimization, and applying the Gittins index to minimizing tail latency in queues. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 16:38:51 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 22:41:05 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Scully",
"Ziv",
""
],
[
"Terenin",
"Alexander",
""
]
] |
2506.11240 | Shai Keidar | Shai Keidar, Shaul Ragimov | Twisted Graded Categories | 74 pages. Comments are welcome! | null | null | null | math.AT math.CT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Given a presentably symmetric monoidal $\infty$-category $\mathcal{C}$ and an $\mathbb{E}_{\infty}$-monoid $M$, we introduce and classify twisted graded categories, which generalize the Day convolution structure on $\mathrm{Fun}(M, \mathcal{C})$. These are characterized by a braiding encoded in symmetric group actions on tensor powers, whose character we show depends only on the $\mathbb{T}$-equivariant monoidal dimension. We analyze the $\mathbb{T}$-action on the dimension of invertible objects and identify it with the $\mathbb{T}$-transfer map. Finally, we compute braiding characters in examples arising from higher cyclotomic extensions, such as the $(\mathbb{S}, n+1)$-oriented extension of $\mathrm{Mod}_{En}^{\wedge}$ at all primes and heights, and of the cyclotomic closure of $\mathrm{Vect}^n$ at low heights. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 19:18:05 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 19:14:14 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Keidar",
"Shai",
""
],
[
"Ragimov",
"Shaul",
""
]
] |
2506.11247 | Leonid A. Levin | Leonid A. Levin | "Pairs and Squares" Periodic Table | 2 pages | null | null | null | physics.ed-ph physics.chem-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | I present a new "Pairs and Squares" rendering of the Periodic Table. It takes advantage of the number of orbitals at each atomic energy level being a whole square. This makes the table very regular and intuitive in contrast with its currently used presentations. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 19:36:55 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Levin",
"Leonid A.",
""
]
] |
2506.11256 | Gabriel Fernando Rodr\'iguez Ruiz | Gabriel F. Rodr\'iguez Ruiz, Juan Herrera Mateos, Leandro Tosi, Christoph Strunk, Carlos Balseiro and Liliana Arrachea | Antichiral edge states and Bogoliubov Fermi surfaces in a two-dimensional proximity-induced superconductor | null | null | null | null | cond-mat.mes-hall cond-mat.supr-con | http://creativecommons.org/licenses/by/4.0/ | We show that a magnetic field parallel to the plane of a two-dimensional electron gas with Rashba spin orbit coupling in proximity to a superconductor leads to a topological phase in coexistence with a single pair of Bogoliubov Fermi surfaces. This phase hosts antichiral edge states of co-propagating Majorana fermions and are spatially localized at the opposite edges of the sample, perpendicular to the magnetic field. We discuss the characteristic signatures in the current-phase relation of a Josephson junction formed by two reservoirs in the topological phase. | [
{
"version": "v1",
"created": "Thu, 12 Jun 2025 20:00:59 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 14:07:00 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Ruiz",
"Gabriel F. Rodríguez",
""
],
[
"Mateos",
"Juan Herrera",
""
],
[
"Tosi",
"Leandro",
""
],
[
"Strunk",
"Christoph",
""
],
[
"Balseiro",
"Carlos",
""
],
[
"Arrachea",
"Liliana",
""
]
] |
2506.11555 | Yu Wang | Yu Wang, Shiwan Zhao, Zhihu Wang, Yubo Zhang, Xicheng Zhang, Zhengfan Wang, Heyuan Huang, Ming Fan, Ting Liu | RAG+: Enhancing Retrieval-Augmented Generation with Application-Aware Reasoning | null | null | null | null | cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The integration of external knowledge through Retrieval-Augmented Generation (RAG) has become foundational in enhancing large language models (LLMs) for knowledge-intensive tasks. However, existing RAG paradigms often overlook the cognitive step of applying knowledge, leaving a gap between retrieved facts and task-specific reasoning. In this work, we introduce RAG+, a principled and modular extension that explicitly incorporates application-aware reasoning into the RAG pipeline. RAG+ constructs a dual corpus consisting of knowledge and aligned application examples, created either manually or automatically, and retrieves both jointly during inference. This design enables LLMs not only to access relevant information but also to apply it within structured, goal-oriented reasoning processes. Experiments across mathematical, legal, and medical domains, conducted on multiple models, demonstrate that RAG+ consistently outperforms standard RAG variants, achieving average improvements of 3-5%, and peak gains up to 7.5% in complex scenarios. By bridging retrieval with actionable application, RAG+ advances a more cognitively grounded framework for knowledge integration, representing a step toward more interpretable and capable LLMs. | [
{
"version": "v1",
"created": "Fri, 13 Jun 2025 08:06:49 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 05:50:06 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Wang",
"Yu",
""
],
[
"Zhao",
"Shiwan",
""
],
[
"Wang",
"Zhihu",
""
],
[
"Zhang",
"Yubo",
""
],
[
"Zhang",
"Xicheng",
""
],
[
"Wang",
"Zhengfan",
""
],
[
"Huang",
"Heyuan",
""
],
[
"Fan",
"Ming",
""
],
[
"Liu",
"Ting",
""
]
] |
2506.11558 | Bo-Cheng Chiu | Bo-Cheng Chiu, Jen-Jee Chen, Yu-Chee Tseng and Feng-Chi Chen | DaMO: A Data-Efficient Multimodal Orchestrator for Temporal Reasoning with Video LLMs | I would like to request the withdrawal of this submission because the current version contains significant errors and incomplete results. I intend to revise the manuscript thoroughly before resubmitting. I apologize for the oversight and appreciate your understanding | null | null | null | cs.CV cs.AI cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Large Language Models (LLMs) have recently been extended to the video domain, enabling sophisticated video-language understanding. However, existing Video LLMs often exhibit limitations in fine-grained temporal reasoning, restricting their ability to precisely attribute responses to specific video moments, especially under constrained supervision. We introduce DaMO, a data-efficient Video LLM explicitly designed for accurate temporal reasoning and multimodal understanding. At its core, the proposed Temporal-aware Fuseformer employs a hierarchical dual-stream architecture that progressively captures temporal dynamics within each modality and effectively fuses complementary visual and audio information. To further enhance computational efficiency, DaMO integrates a global residual that reduces spatial redundancy while preserving essential semantic details. We train DaMO via a structured four-stage progressive training paradigm, incrementally equipping the model with multimodal alignment, semantic grounding, and temporal reasoning capabilities. This work also contributes multiple datasets augmented from existing ones with GPT-generated temporally grounded QA pairs for tasks requiring temporal supervision. Comprehensive experiments on temporal grounding and video QA benchmarks demonstrate that DaMO consistently surpasses prior methods, particularly in tasks demanding precise temporal alignment and reasoning. Our work establishes a promising direction for data-efficient video-language modeling. | [
{
"version": "v1",
"created": "Fri, 13 Jun 2025 08:13:05 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 11:59:30 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Chiu",
"Bo-Cheng",
""
],
[
"Chen",
"Jen-Jee",
""
],
[
"Tseng",
"Yu-Chee",
""
],
[
"Chen",
"Feng-Chi",
""
]
] |
2506.11855 | Riccardo Castellano | Riccardo Castellano, Vasco Cavina, Mart\'i Perarnau-Llobet, Pavel Sekatski, Vittorio Giovannetti | Exact requirements for battery-assisted qubit gates | 20 pages, 1 figure | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | We consider the implementation of a unitary gate on a qubit system S via a global energy-preserving operation acting on S and an auxiliary system B that can be seen as a battery. We derive a simple, asymptotically exact expression for the implementation error as a function of the battery state, which we refer to as the it Unitary Defect. Remarkably, this quantity is independent of the specific gate being implemented, highlighting a universal property of the battery itself. We show that minimizing the unitary defect, under given physical constraints on the battery state, is mathematically equivalent to solving a Lagrangian optimization problem, often corresponding to finding the ground state of a one-dimensional quantum system. Using this mapping, we identify optimal battery states that achieve the highest precision under constraints on energy, squared energy, number of levels and Quantum Fisher Information. Overall, our results provide an efficient method for establishing bounds on the physical requirements needed to implement a unitary gate via energy-preserving operations and for determining the corresponding optimal protocols. | [
{
"version": "v1",
"created": "Fri, 13 Jun 2025 15:01:34 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 14:46:50 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Castellano",
"Riccardo",
""
],
[
"Cavina",
"Vasco",
""
],
[
"Perarnau-Llobet",
"Martí",
""
],
[
"Sekatski",
"Pavel",
""
],
[
"Giovannetti",
"Vittorio",
""
]
] |
2506.11903 | Raphael Scheible-Schmitt | Raphael Scheible-Schmitt and Johann Frei | GeistBERT: Breathing Life into German NLP | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Advances in transformer-based language models have highlighted the benefits of language-specific pre-training on high-quality corpora. In this context, German NLP stands to gain from updated architectures and modern datasets tailored to the linguistic characteristics of the German language. GeistBERT seeks to improve German language processing by incrementally training on a diverse corpus and optimizing model performance across various NLP tasks. It was pre-trained using fairseq with standard hyperparameters, initialized from GottBERT weights, and trained on a large-scale German corpus using Whole Word Masking (WWM). Based on the pre-trained model, we derived extended-input variants using Nystr\"omformer and Longformer architectures with support for sequences up to 8k tokens. While these long-context models were not evaluated on dedicated long-context benchmarks, they are included in our release. We assessed all models on NER (CoNLL 2003, GermEval 2014) and text classification (GermEval 2018 fine/coarse, 10kGNAD) using $F_1$ score and accuracy. The GeistBERT models achieved strong performance, leading all tasks among the base models and setting a new state-of-the-art (SOTA). Notably, the base models outperformed larger models in several tasks. To support the German NLP research community, we are releasing GeistBERT under the MIT license. | [
{
"version": "v1",
"created": "Fri, 13 Jun 2025 15:53:17 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 23:06:09 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Jun 2025 12:31:06 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Scheible-Schmitt",
"Raphael",
""
],
[
"Frei",
"Johann",
""
]
] |
2506.12006 | Navodini Wijethilake | Navodini Wijethilake, Reuben Dorent, Marina Ivory, Aaron Kujawa, Stefan Cornelissen, Patrick Langenhuizen, Mohamed Okasha, Anna Oviedova, Hexin Dong, Bogyeong Kang, Guillaume Sall\'e, Luyi Han, Ziyuan Zhao, Han Liu, Tao Yang, Shahad Hardan, Hussain Alasmawi, Santosh Sanjeev, Yuzhou Zhuang, Satoshi Kondo, Maria Baldeon Calisto, Shaikh Muhammad Uzair Noman, Cancan Chen, Ipek Oguz, Rongguo Zhang, Mina Rezaei, Susana K. Lai-Yuen, Satoshi Kasai, Chih-Cheng Hung, Mohammad Yaqub, Lisheng Wang, Benoit M. Dawant, Cuntai Guan, Ritse Mann, Vincent Jaouen, Ji-Wung Han, Li Zhang, Jonathan Shapey, Tom Vercauteren | crossMoDA Challenge: Evolution of Cross-Modality Domain Adaptation Techniques for Vestibular Schwannoma and Cochlea Segmentation from 2021 to 2023 | null | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The cross-Modality Domain Adaptation (crossMoDA) challenge series, initiated in 2021 in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), focuses on unsupervised cross-modality segmentation, learning from contrast-enhanced T1 (ceT1) and transferring to T2 MRI. The task is an extreme example of domain shift chosen to serve as a meaningful and illustrative benchmark. From a clinical application perspective, it aims to automate Vestibular Schwannoma (VS) and cochlea segmentation on T2 scans for more cost-effective VS management. Over time, the challenge objectives have evolved to enhance its clinical relevance. The challenge evolved from using single-institutional data and basic segmentation in 2021 to incorporating multi-institutional data and Koos grading in 2022, and by 2023, it included heterogeneous routine data and sub-segmentation of intra- and extra-meatal tumour components. In this work, we report the findings of the 2022 and 2023 editions and perform a retrospective analysis of the challenge progression over the years. The observations from the successive challenge contributions indicate that the number of outliers decreases with an expanding dataset. This is notable since the diversity of scanning protocols of the datasets concurrently increased. The winning approach of the 2023 edition reduced the number of outliers on the 2021 and 2022 testing data, demonstrating how increased data heterogeneity can enhance segmentation performance even on homogeneous data. However, the cochlea Dice score declined in 2023, likely due to the added complexity from tumour sub-annotations affecting overall segmentation performance. While progress is still needed for clinically acceptable VS segmentation, the plateauing performance suggests that a more challenging cross-modal task may better serve future benchmarking. | [
{
"version": "v1",
"created": "Fri, 13 Jun 2025 17:56:39 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 14:00:00 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Wijethilake",
"Navodini",
""
],
[
"Dorent",
"Reuben",
""
],
[
"Ivory",
"Marina",
""
],
[
"Kujawa",
"Aaron",
""
],
[
"Cornelissen",
"Stefan",
""
],
[
"Langenhuizen",
"Patrick",
""
],
[
"Okasha",
"Mohamed",
""
],
[
"Oviedova",
"Anna",
""
],
[
"Dong",
"Hexin",
""
],
[
"Kang",
"Bogyeong",
""
],
[
"Sallé",
"Guillaume",
""
],
[
"Han",
"Luyi",
""
],
[
"Zhao",
"Ziyuan",
""
],
[
"Liu",
"Han",
""
],
[
"Yang",
"Tao",
""
],
[
"Hardan",
"Shahad",
""
],
[
"Alasmawi",
"Hussain",
""
],
[
"Sanjeev",
"Santosh",
""
],
[
"Zhuang",
"Yuzhou",
""
],
[
"Kondo",
"Satoshi",
""
],
[
"Calisto",
"Maria Baldeon",
""
],
[
"Noman",
"Shaikh Muhammad Uzair",
""
],
[
"Chen",
"Cancan",
""
],
[
"Oguz",
"Ipek",
""
],
[
"Zhang",
"Rongguo",
""
],
[
"Rezaei",
"Mina",
""
],
[
"Lai-Yuen",
"Susana K.",
""
],
[
"Kasai",
"Satoshi",
""
],
[
"Hung",
"Chih-Cheng",
""
],
[
"Yaqub",
"Mohammad",
""
],
[
"Wang",
"Lisheng",
""
],
[
"Dawant",
"Benoit M.",
""
],
[
"Guan",
"Cuntai",
""
],
[
"Mann",
"Ritse",
""
],
[
"Jaouen",
"Vincent",
""
],
[
"Han",
"Ji-Wung",
""
],
[
"Zhang",
"Li",
""
],
[
"Shapey",
"Jonathan",
""
],
[
"Vercauteren",
"Tom",
""
]
] |
2506.12023 | Paolo Roselli | Paolo Roselli, Michel Willem | On the validity of the Radon-Nikodym Theorem | null | null | null | null | math.GM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper presents a new general formulation of the Radon-Nikodym theorem in the setting of abstract measure theory. We introduce the notion of weak localizability for a measure and show that this property is both necessary and sufficient for the validity of a Radon-Nikodym-type representation under a natural compatibility relation between measures. The proof relies solely on elementary tools, such as Markov's inequality and the monotone convergence theorem. In addition to establishing the main result, we provide a constructive approach to envelope functions for families of non-negative measurable functions supported on sets of finite measure. | [
{
"version": "v1",
"created": "Mon, 12 May 2025 12:03:00 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 10:29:04 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Roselli",
"Paolo",
""
],
[
"Willem",
"Michel",
""
]
] |
2506.12374 | Wenbo Li | Wenbo Li, Shiyi Wang, Yiteng Chen, Huiping Zhuang and Qingyao Wu | AntiGrounding: Lifting Robotic Actions into VLM Representation Space for Decision Making | submitted to NeurIPS 2025 | null | null | null | cs.RO cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Vision-Language Models (VLMs) encode knowledge and reasoning capabilities for robotic manipulation within high-dimensional representation spaces. However, current approaches often project them into compressed intermediate representations, discarding important task-specific information such as fine-grained spatial or semantic details. To address this, we propose AntiGrounding, a new framework that reverses the instruction grounding process. It lifts candidate actions directly into the VLM representation space, renders trajectories from multiple views, and uses structured visual question answering for instruction-based decision making. This enables zero-shot synthesis of optimal closed-loop robot trajectories for new tasks. We also propose an offline policy refinement module that leverages past experience to enhance long-term performance. Experiments in both simulation and real-world environments show that our method outperforms baselines across diverse robotic manipulation tasks. | [
{
"version": "v1",
"created": "Sat, 14 Jun 2025 07:11:44 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 10:01:18 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Li",
"Wenbo",
""
],
[
"Wang",
"Shiyi",
""
],
[
"Chen",
"Yiteng",
""
],
[
"Zhuang",
"Huiping",
""
],
[
"Wu",
"Qingyao",
""
]
] |
2506.12398 | Michael \'O Ceallaigh | Lisa Orloff Clark, Michael \'O Ceallaigh, Hung Pham | Generalised Twisted Groupoids and their C*-algebras | 21 pages | null | null | null | math.OA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We consider a locally compact Hausdorff groupoid $G$, and twist by a more general locally compact Hausdorff abelian group $\Gamma$ rather than the complex unit circle $\mathbb{T}$. We investigate the construction of $C^*$-algebras in analogue to the usual twisted groupoid $C^*$-algebras, and we show that, in fact, any $\Gamma$-twisted groupoid $C^*$-algebra is isomorphic to a usual twisted groupoid $C^*$-algebra. | [
{
"version": "v1",
"created": "Sat, 14 Jun 2025 08:24:02 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 23:21:35 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Clark",
"Lisa Orloff",
""
],
[
"Ceallaigh",
"Michael Ó",
""
],
[
"Pham",
"Hung",
""
]
] |
2506.12884 | Zikang Ge | Zikang Ge, Jinhao Hu, Shengyuan Peng, Wei Kang, Xiaofei Shen, Yanbo Xie, and Jianming Xue | Ion Track Formation via Electric-Field-Enhanced Nanoscale Energy Deposition | null | null | null | null | physics.app-ph cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | High-energy ion irradiation deposits extreme energy in a narrow range (1-10 nm) along ion trajectories in solid through electronic energy loss, producing unique irradiation effects such as ion tracks. However, intrinsic velocity effects impose an upper limit on electronic energy loss that cannot be overcome by adjusting irradiation parameters. We introduce a method using electric fields during irradiation to enhance nanoscale energy deposition by accelerating ion-excited electrons within sub-picosecond timescales.Our extended thermal spike model quantitatively describes this enhancement and predicts a significant reduction in the electronic energy loss required for ion track formation in amorphous SiO2, which is in excellent agreement with experimental observations. This work provides a new approach to control energy deposition during irradiation and boosts the wide application of ion tracks in material modification and nanoengineering to much broader extents. | [
{
"version": "v1",
"created": "Sun, 15 Jun 2025 15:28:22 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 11:36:40 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Ge",
"Zikang",
""
],
[
"Hu",
"Jinhao",
""
],
[
"Peng",
"Shengyuan",
""
],
[
"Kang",
"Wei",
""
],
[
"Shen",
"Xiaofei",
""
],
[
"Xie",
"Yanbo",
""
],
[
"Xue",
"Jianming",
""
]
] |
2506.13033 | Joao V. Cavalcanti | Joao V. Cavalcanti and Laurent Lessard and Ashia C. Wilson | Adaptive Acceleration Without Strong Convexity Priors Or Restarts | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | Accelerated optimization methods are typically parametrized by the Lipschitz smoothness and strong convexity parameters of a given problem, which are rarely known in practice. While the Lipschitz smoothness parameter L can be easily estimated by backtracking, directly estimating (i.e., without restarts) the strong convexity parameter m remains an open problem. This paper addresses this challenge by proposing NAG-free, an optimization method that achieves acceleration while estimating m online without priors or restarts. Our method couples Nesterov's accelerated gradient (NAG) method with an inexpensive estimator that does not require additional computation, only storage of one more iterate and gradient. We prove that in the canonical setting of the open problem, NAG-free converges globally at least as fast as gradient descent and locally at an accelerated rate, without resorting to m priors or restarts. We also introduce a complementary estimator for L, symmetric to the m estimator. We provide extensive empirical evidence that this NAG-free heuristic with L and m estimators often achieves acceleration in practice, outperforming restart schemes and traditional accelerated methods parametrized by offline estimates of L and m. Our experiments also suggest that NAG-free might be effective even for nonconvex problems when combined with restarts. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 01:54:18 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 18:40:42 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Cavalcanti",
"Joao V.",
""
],
[
"Lessard",
"Laurent",
""
],
[
"Wilson",
"Ashia C.",
""
]
] |
2506.13040 | Hanz Cuevas Velasquez | Hanz Cuevas-Velasquez, Anastasios Yiannakidis, Soyong Shin, Giorgio Becherini, Markus H\"oschle, Joachim Tesch, Taylor Obersat, Tsvetelina Alexiadis, Michael J. Black | MAMMA: Markerless & Automatic Multi-Person Motion Action Capture | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | We present MAMMA, a markerless motion-capture pipeline that accurately recovers SMPL-X parameters from multi-view video of two-person interaction sequences. Traditional motion-capture systems rely on physical markers. Although they offer high accuracy, their requirements of specialized hardware, manual marker placement, and extensive post-processing make them costly and time-consuming. Recent learning-based methods attempt to overcome these limitations, but most are designed for single-person capture, rely on sparse keypoints, or struggle with occlusions and physical interactions. In this work, we introduce a method that predicts dense 2D surface landmarks conditioned on segmentation masks, enabling person-specific correspondence estimation even under heavy occlusion. We employ a novel architecture that exploits learnable queries for each landmark. We demonstrate that our approach can handle complex person--person interaction and offers greater accuracy than existing methods. To train our network, we construct a large, synthetic multi-view dataset combining human motions from diverse sources, including extreme poses, hand motions, and close interactions. Our dataset yields high-variability synthetic sequences with rich body contact and occlusion, and includes SMPL-X ground-truth annotations with dense 2D landmarks. The result is a system capable of capturing human motion without the need for markers. Our approach offers competitive reconstruction quality compared to commercial marker-based motion-capture solutions, without the extensive manual cleanup. Finally, we address the absence of common benchmarks for dense-landmark prediction and markerless motion capture by introducing two evaluation settings built from real multi-view sequences. We will release our dataset, benchmark, method, training code, and pre-trained model weights for research purposes. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 02:04:51 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 15:25:06 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Cuevas-Velasquez",
"Hanz",
""
],
[
"Yiannakidis",
"Anastasios",
""
],
[
"Shin",
"Soyong",
""
],
[
"Becherini",
"Giorgio",
""
],
[
"Höschle",
"Markus",
""
],
[
"Tesch",
"Joachim",
""
],
[
"Obersat",
"Taylor",
""
],
[
"Alexiadis",
"Tsvetelina",
""
],
[
"Black",
"Michael J.",
""
]
] |
2506.13097 | Ziqing Zhou | Ziqing Zhou, Yurui Pan, Lidong Wang, Wenbing Zhu, Mingmin Chi, Dong Wu, Bo Peng | Pro-AD: Learning Comprehensive Prototypes with Prototype-based Constraint for Multi-class Unsupervised Anomaly Detection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Prototype-based reconstruction methods for unsupervised anomaly detection utilize a limited set of learnable prototypes which only aggregates insufficient normal information, resulting in undesirable reconstruction. However, increasing the number of prototypes may lead to anomalies being well reconstructed through the attention mechanism, which we refer to as the "Soft Identity Mapping" problem. In this paper, we propose Pro-AD to address these issues and fully utilize the prototypes to boost the performance of anomaly detection. Specifically, we first introduce an expanded set of learnable prototypes to provide sufficient capacity for semantic information. Then we employ a Dynamic Bidirectional Decoder which integrates the process of the normal information aggregation and the target feature reconstruction via prototypes, with the aim of allowing the prototypes to aggregate more comprehensive normal semantic information from different levels of the image features and the target feature reconstruction to not only utilize its contextual information but also dynamically leverage the learned comprehensive prototypes. Additionally, to prevent the anomalies from being well reconstructed using sufficient semantic information through the attention mechanism, Pro-AD introduces a Prototype-based Constraint that applied within the target feature reconstruction process of the decoder, which further improves the performance of our approach. Extensive experiments on multiple challenging benchmarks demonstrate that our Pro-AD achieve state-of-the-art performance, highlighting its superior robustness and practical effectiveness for Multi-class Unsupervised Anomaly Detection task. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 05:04:12 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 09:23:19 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Jun 2025 02:59:10 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Zhou",
"Ziqing",
""
],
[
"Pan",
"Yurui",
""
],
[
"Wang",
"Lidong",
""
],
[
"Zhu",
"Wenbing",
""
],
[
"Chi",
"Mingmin",
""
],
[
"Wu",
"Dong",
""
],
[
"Peng",
"Bo",
""
]
] |
2506.13357 | Ilya Barmak | Ilya Barmak, Subham Pal, Alexander Gelfgat, Neima Brauner | Two-phase stratified MHD flows in wide rectangular ducts: analytical and numerical solutions | 37 pages, 20 figures | null | null | null | physics.flu-dyn | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study explores the effects of a non-conductive gas layer flowing concurrently with a conductive liquid on the two-phase flow characteristics in wide horizontal ducts under a constant vertical magnetic field. To this end, analytical solutions for the velocity profile and induced magnetic field are presented for laminar gas-liquid stratified magnetohydrodynamic (MHD) flow between two infinite plates of various conductivities. The contributions of the Lorentz force and wall shear stresses to the pressure gradient are examined. To the best of our knowledge, it is shown for the first time that, unlike the single-phase Hartmann flow, the velocity profiles in two-phase flow differ significantly depending on whether the bottom wall is conducting or insulating. In the case of an insulating bottom wall, the gas lubrication effect and potential pumping power savings are significantly greater, regardless of the magnetic Reynolds number. This conclusion also holds for gas-liquid MHD flows in rectangular ducts with finite width-to-height aspect ratios. To assess the applicability of the Two-Plate (TP) model to wide ducts, numerical solutions of the two-dimensional problem are used to investigate the influence of side walls on the two-phase flow characteristics, considering various combinations of bottom and side wall conductivities. In all cases, the results for high aspect ratios converge to the analytical solution obtained from the TP model with the same bottom wall conductivity. However, the influence of insulating side walls remains significant even at large aspect ratios when the bottom wall is conducting. Unexpectedly, in such cases, the change in the induced magnetic field due to the presence of side walls has a dramatic effect on the velocity profile, leading to a reduced pressure gradient compared to that predicted by the TP model. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 10:55:46 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 13:12:23 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Barmak",
"Ilya",
""
],
[
"Pal",
"Subham",
""
],
[
"Gelfgat",
"Alexander",
""
],
[
"Brauner",
"Neima",
""
]
] |
2506.13459 | Rebecca Barrett | R. Barrett (1), C. D. Tremblay (2, 3 and 4), B. Addison (5 and 6), D. C. Price (7 and 8), J. A. Green (8) ((1) University of Southern Queensland, (2) SETI Institute, (3) Berkeley SETI Research Center, (4) CSIRO Astronomy and Space Science, (5) University of Southern Queensland Centre for Astrophysics, (6) Swinburne University of Technology Centre for Astrophysics and Supercomputing, (7) International Centre for Radio Astronomy Research, (8) SKA Observatory Science Operations Centre) | Breakthrough Listen: A Technosignature Search Around 27 Eclipsing Exoplanets Selected from the Transiting Exoplanet Survey Satellite Catalogue | Accepted for publication in Publications of the Astronomical Society of Australia (PASA). 9 pages, 5 figures, 1 table | null | 10.1017/pasa.2025.10060 | null | astro-ph.EP astro-ph.IM | http://creativecommons.org/licenses/by/4.0/ | Here we analyse the archival data for a set of 27 Transiting Exoplanet Survey Satellite (TESS) Targets of Interest (TOIs) in search for artificially generated radio signals, or 'technosignatures', interrupted by occultation. Exoplanetary eclipses are notable events to observe in the search for technosignatures, as they mark the geometrical alignment of the target, its host star, and Earth. During an eclipse event, any signal emanating from the target of interest should cease for the duration of the eclipse, and resume after the line-of-sight has been restored. Target observations were made by Breakthrough Listen (BL) using Murriyang, the CSIRO Parkes 64-m radio telescope, coupled with the Ultra-wide Low frequency (UWL) receiver covering a continuous range of frequencies spanning 704-4032 MHz inclusive. Each target was observed in a pattern consisting of six back-to-back 5-minute source and reference sky positions for comparison during data analysis. We performed a Doppler search for narrowband signals with a minimum signal-to-noise (S/N) ratio of 10, a minimum drift rate of $\pm0.1$ Hz/s, and a maximum drift rate of $\pm4.0$ Hz/s using the turboSETI pipeline. In the analysis of 1,954,880 signals, 14,639 passed automated radio interference filters where each event was presented as a set of stacked dynamic spectra. Despite manually inspecting each diagram for a signal of interest, all events were attributed to terrestrial radio frequency interference (RFI). | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 13:15:11 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Barrett",
"R.",
"",
"University of Southern Queensland"
],
[
"Tremblay",
"C. D.",
"",
"2, 3 and 4"
],
[
"Addison",
"B.",
"",
"5 and 6"
],
[
"Price",
"D. C.",
"",
"7 and 8"
],
[
"Green",
"J. A.",
"",
"SKA Observatory Science Operations Centre"
]
] |
2506.13699 | Wadia Mansouri | T. A\"it Aissa, S. El Bourkadi, M. W. Mansouri and SM. Sbai | Eight-dimensional non completely reducible symplectic Lie algebras | null | null | null | null | math.SG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A non completely reducible symplectic Lie algebra is a symplectic Lie algebra which cannot be symplectically reduced to the trivial symplectic Lie algebra. Our aim is to provide a complete classification, up to symplectomorphism of non completely reducible symplectic Lie algebras in dimensions $n \leq 8$ and, furthermore, to provide a complete description of symplectic Lie algebras admitting one-dimensional isotropic ideals. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 17:05:10 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 17:40:29 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Aissa",
"T. Aït",
""
],
[
"Bourkadi",
"S. El",
""
],
[
"Mansouri",
"M. W.",
""
],
[
"Sbai",
"SM.",
""
]
] |
2506.13876 | Ashoke Sen | Ashoke Sen | Decorating Asymptotically Flat Space-Time with the Moduli Space of String Theory | LaTeX file, 39 pages; v2: added comment on cobordism conjecture in footnote 1 | null | null | null | hep-th gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | N=2, 4 and 8 supersymmetric string theories in four dimensional flat space-time have moduli space of vacua. We argue that starting from a theory where the moduli approach a particular moduli space point A at infinity, we can construct a classical solution that contains an arbitrarily large space-time region where the moduli take values corresponding to any other moduli space point B of our choice to any desired accuracy. Therefore the observables of a theory with a given set of asymptotic values of the moduli will have complete information on the observables for any other asymptotic values of the moduli. Also it is physically impossible for any experiment, performed over a finite time, to determine the asymptotic values of the moduli. We point out the difference between asymptotically flat space-time and asymptotically AdS space-time in this regard and discuss the possible implication of these results for holographic duals of string theories in flat space-time. For N=2 supersymmetric theories, A and B could correspond to compactifications on topologically distinct Calabi-Yau manifolds related by flop or conifold transitions. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 18:00:15 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 07:03:45 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Sen",
"Ashoke",
""
]
] |
2506.13930 | Mateo Restrepo Borrero | Mateo Restrepo Borrero and Khodr Shamseddine | On a Lebesgue-like integral over the Levi-Civita field $\mathcal{R}$ | null | null | null | null | math.FA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Levi-Civita field $\mathcal{R}$ is the smallest non-Archimedean ordered field extension of the real numbers that is real closed and Cauchy complete in the topology induced by the order. In this paper we develop a new theory of integration over $\mathcal{R}$ that generalizes previous work done in the subject while circumventing the fact that not every bounded subset of $\mathcal{R}$ admits either an infimum or a supremum. We define a new family of measurable functions and a new integral over measurable subsets of $\mathcal{R}$ that satisfies some very important results analogous to those of the Lebesgue and Riemann integrals for real-valued functions. In particular, we show that the family of measurable functions forms an algebra that is closed under taking absolute values, that the integral is linear, countably additive and monotone, that the integral of a non-negative function $f:A\to\mathcal{R}$ is zero if and only if $f=0$ almost everywhere in $A$ and that the analogous versions for the uniform convergence theorem and the fundamental theorem of calculus hold true. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 19:16:12 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 22:01:51 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Borrero",
"Mateo Restrepo",
""
],
[
"Shamseddine",
"Khodr",
""
]
] |
2506.13952 | Adrian Ezequiel Rubio Lopez | Adrian E. Rubio Lopez, Felipe Herrera | Detecting zero-point fluctuations with stochastic Brownian oscillators | 5 pages, 3 figures, 12 pages of supplementary material | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | High-quality quantum oscillators are preferred for precision sensing of external physical parameter because if the noise level due to interactions with the environment is too high, metrological information can be lost due to quantum decoherence. On the other hand, stronger interactions with a thermal environment could be seen a resource for new types of metrological schemes. We present a general amplification strategy that enables the detection zero-point fluctuations using low-quality quantum oscillators at finite temperature. We show that by injecting a controllable level of multiplicative frequency noise in a Brownian oscillator, quantum deviations from the virial theorem can be amplified by a parameter proportional to the strength of the frequency noise at constant temperature. As an application, we suggest a scheme in which the virial ratio is used as a witness of the quantum fluctuations of an unknown thermal bath, either by measuring the oscillator energy or the heat current flowing into an ancilla bath. Our work expands the metrological capacity of low-quality oscillators and can enable new measurements of the quantum properties of thermal environments by sensing their zero-point contributions to system variables. | [
{
"version": "v1",
"created": "Mon, 16 Jun 2025 19:45:12 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 16:35:29 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Lopez",
"Adrian E. Rubio",
""
],
[
"Herrera",
"Felipe",
""
]
] |
2506.14230 | Kohei Kobayashi | Kohei Kobayashi | Norm Inequality for Perturbed Quantum Evolutions and Its Application to Grover's Algorithm | 5 pages, 1 figure | null | null | null | quant-ph | http://creativecommons.org/licenses/by-sa/4.0/ | We investigate the impact of coherent control errors on quantum state evolution by deriving a general norm inequality based on Gronwall's lemma. This inequality provides an explicit upper bound on the deviation between an ideal quantum state and one subject to arbitrary coherent perturbations, including both time-dependent and time-independent cases. The framework is broadly applicable, requiring no assumptions about the detailed structure of the perturbation and full dynamics of the quantum system. We apply this approach to analyze the robustness of Grover's search algorithm in the presence of coherent errors. Our results characterizes quantitative scaling relations between the error strength, algorithm runtime, and success probability, offering practical guidelines for the design of resilient quantum protocols. We further compare the effects of time-dependent and time-independent perturbations, showing the distinctive ways in which coherent errors accumulate in quantum dynamics. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 06:35:51 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 11:51:50 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Jun 2025 06:26:16 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Kobayashi",
"Kohei",
""
]
] |
2506.14487 | Paolo Franceschi | Paolo Franceschi, Marco Faroni, Stefano Baraldo, Anna Valente | ros2 fanuc interface: Design and Evaluation of a Fanuc CRX Hardware Interface in ROS2 | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper introduces the ROS2 control and the Hardware Interface (HW) integration for the Fanuc CRX- robot family. It explains basic implementation details and communication protocols, and its integration with the Moveit2 motion planning library. We conducted a series of experiments to evaluate relevant performances in the robotics field. We tested the developed ros2_fanuc_interface for four relevant robotics cases: step response, trajectory tracking, collision avoidance integrated with Moveit2, and dynamic velocity scaling, respectively. Results show that, despite a non-negligible delay between command and feedback, the robot can track the defined path with negligible errors (if it complies with joint velocity limits), ensuring collision avoidance. Full code is open source and available at https://github.com/paolofrance/ros2_fanuc_interface. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 13:08:32 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 13:40:12 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Franceschi",
"Paolo",
""
],
[
"Faroni",
"Marco",
""
],
[
"Baraldo",
"Stefano",
""
],
[
"Valente",
"Anna",
""
]
] |
2506.14662 | Young-Ho Cho | Young-ho Cho, Min-Seung Ko, Hao Zhu | PGLib-CO2: A Power Grid Library for Computing and Optimizing Carbon Emissions | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | A sustainable electricity infrastructure requires the explicit integration of carbon emissions into power system modeling and optimization paradigms. However, existing open-source datasets for power system R&D lack generator-level carbon emission profiling, limiting the ability to benchmark and compare various carbon-aware grid operational strategies. To address this gap, this work introduces PGLib-CO2, an open-source extension to the widely adopted PGLib-OPF test case library. PGLib-CO2 enriches standard network cases with CO2 and CO2-equivalent emission intensity factors by expanding the fuel-type categorization used by PGLib-OPF, attaining a realistic generator-level carbon profiling. It is also packaged for both Python's pandapower and Julia's PowerModels.jl, for a seamless, user-friendly integration of emission modeling into grid computation and optimization tasks. The dataset produced by PGLib-CO2 can support grid-based carbon accounting, emission metric evaluation, and integration into AC optimal power flow (OPF) and optimal load shifting (OLS) formulations. We demonstrate PGLib-CO2's utility through case studies that quantify cost-emission trade-offs and optimize a carbon-aware objective function. By standardizing carbon-enhanced test cases, PGLib-CO2 provides an open-source, reproducible foundation for benchmarking carbon-aware computation, facilitating future research in sustainable power system operation. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:52:47 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 19:41:39 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Cho",
"Young-ho",
""
],
[
"Ko",
"Min-Seung",
""
],
[
"Zhu",
"Hao",
""
]
] |
2506.14667 | Matt Poyser | Matt Poyser and Toby P. Breckon | DDS-NAS: Dynamic Data Selection within Neural Architecture Search via On-line Hard Example Mining applied to Image Classification | 27 single-column pages, 8 figures, to be published in Pattern Recognition | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In order to address the scalability challenge within Neural Architecture Search (NAS), we speed up NAS training via dynamic hard example mining within a curriculum learning framework. By utilizing an autoencoder that enforces an image similarity embedding in latent space, we construct an efficient kd-tree structure to order images by furthest neighbour dissimilarity in a low-dimensional embedding. From a given query image from our subsample dataset, we can identify the most dissimilar image within the global dataset in logarithmic time. Via curriculum learning, we then dynamically re-formulate an unbiased subsample dataset for NAS optimisation, upon which the current NAS solution architecture performs poorly. We show that our DDS-NAS framework speeds up gradient-based NAS strategies by up to 27x without loss in performance. By maximising the contribution of each image sample during training, we reduce the duration of a NAS training cycle and the number of iterations required for convergence. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:58:10 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 01:31:33 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Poyser",
"Matt",
""
],
[
"Breckon",
"Toby P.",
""
]
] |
2506.14677 | Yingchao Li | Yingchao Li | Human-Centered Editable Speech-to-Sign-Language Generation via Streaming Conformer-Transformer and Resampling Hook | null | null | null | null | cs.HC cs.AI | http://creativecommons.org/licenses/by/4.0/ | Existing end-to-end sign-language animation systems suffer from low naturalness, limited facial/body expressivity, and no user control. We propose a human-centered, real-time speech-to-sign animation framework that integrates (1) a streaming Conformer encoder with an autoregressive Transformer-MDN decoder for synchronized upper-body and facial motion generation, (2) a transparent, editable JSON intermediate representation empowering deaf users and experts to inspect and modify each sign segment, and (3) a human-in-the-loop optimization loop that refines the model based on user edits and ratings. Deployed on Unity3D, our system achieves a 13 ms average frame-inference time and a 103 ms end-to-end latency on an RTX 4070. Our key contributions include the design of a JSON-centric editing mechanism for fine-grained sign-level personalization and the first application of an MDN-based feedback loop for continuous model adaptation. This combination establishes a generalizable, explainable AI paradigm for user-adaptive, low-latency multimodal systems. In studies with 20 deaf signers and 5 professional interpreters, we observe a +13 point SUS improvement, 6.7 point reduction in cognitive load, and significant gains in naturalness and trust (p $<$ .001) over baselines. This work establishes a scalable, explainable AI paradigm for accessible sign-language technologies. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:08:48 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 13:11:34 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Li",
"Yingchao",
""
]
] |
2506.14860 | Rui Yang | Yizhi Liang, Jun Tao and Rui Yang | Black holes in $f(R,T)$ gravity coupled with Euler-Heisenberg electrodynamics | 27 pages, 5 figures | null | null | CTP-SCU/2025014 | gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We investigate the scenario of black holes coupled with the Euler-Heisenberg nonlinear electromagnetic field in the framework of $f(R,T)$ gravity. The black hole solutions for electrically charged, magnetically charged and the dyonic case are separately analyzed, and we discuss the scalar curvature and the energy conditions of the black hole spacetime. In the magnetic charge solution, the $f(R,T)$ correction appears in the $1/r^6$ Euler-Heisenberg electromagnetic field correction term, while the electrically charged solution includes an $r^2$ term, with the curvature of the spacetime determined by the $f(R,T)$ coupling parameter. The dyonic solution is obtained through vacuum polarization quantum electrodynamics corrections, where electromagnetic duality is broken and the solution exists higher-order correction terms. The relationship between the event horizon and charge of dyonic extreme black holes is studied. Furthermore, we investigate the effective metric, photon trajectories, and innermost stable circular orbit under nonlinear electromagnetic effects, providing images of photon geodesic for varying electric and magnetic charge strengths and $f(R,T)$ coupling parameter. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 13:38:52 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 13:01:35 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Liang",
"Yizhi",
""
],
[
"Tao",
"Jun",
""
],
[
"Yang",
"Rui",
""
]
] |
2506.14920 | Daniel Ruiz | Alejandro Giraldo and Daniel Ruiz and Mariano Caruso and Javier Mancilla and Guido Bellomo | Q2SAR: A Quantum Multiple Kernel Learning Approach for Drug Discovery | null | null | null | null | quant-ph cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Quantitative Structure-Activity Relationship (QSAR) modeling is a cornerstone of computational drug discovery. This research demonstrates the successful application of a Quantum Multiple Kernel Learning (QMKL) framework to enhance QSAR classification, showing a notable performance improvement over classical methods. We apply this methodology to a dataset for identifying DYRK1A kinase inhibitors. The workflow involves converting SMILES representations into numerical molecular descriptors, reducing dimensionality via Principal Component Analysis (PCA), and employing a Support Vector Machine (SVM) trained on an optimized combination of multiple quantum and classical kernels. By benchmarking the QMKL-SVM against a classical Gradient Boosting model, we show that the quantum-enhanced approach achieves a superior AUC score, highlighting its potential to provide a quantum advantage in challenging cheminformatics classification tasks. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 19:00:47 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 15:57:50 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Giraldo",
"Alejandro",
""
],
[
"Ruiz",
"Daniel",
""
],
[
"Caruso",
"Mariano",
""
],
[
"Mancilla",
"Javier",
""
],
[
"Bellomo",
"Guido",
""
]
] |
2506.14950 | Daqian Shao | Daqian Shao, Ashkan Soleymani, Francesco Quinzan, Marta Kwiatkowska | Double Machine Learning for Conditional Moment Restrictions: IV Regression, Proximal Causal Learning and Beyond | null | null | null | null | stat.ML cs.LG stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Solving conditional moment restrictions (CMRs) is a key problem considered in statistics, causal inference, and econometrics, where the aim is to solve for a function of interest that satisfies some conditional moment equalities. Specifically, many techniques for causal inference, such as instrumental variable (IV) regression and proximal causal learning (PCL), are CMR problems. Most CMR estimators use a two-stage approach, where the first-stage estimation is directly plugged into the second stage to estimate the function of interest. However, naively plugging in the first-stage estimator can cause heavy bias in the second stage. This is particularly the case for recently proposed CMR estimators that use deep neural network (DNN) estimators for both stages, where regularisation and overfitting bias is present. We propose DML-CMR, a two-stage CMR estimator that provides an unbiased estimate with fast convergence rate guarantees. We derive a novel learning objective to reduce bias and develop the DML-CMR algorithm following the double/debiased machine learning (DML) framework. We show that our DML-CMR estimator can achieve the minimax optimal convergence rate of $O(N^{-1/2})$ under parameterisation and mild regularity conditions, where $N$ is the sample size. We apply DML-CMR to a range of problems using DNN estimators, including IV regression and proximal causal learning on real-world datasets, demonstrating state-of-the-art performance against existing CMR estimators and algorithms tailored to those problems. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 20:00:34 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 18:27:16 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Shao",
"Daqian",
""
],
[
"Soleymani",
"Ashkan",
""
],
[
"Quinzan",
"Francesco",
""
],
[
"Kwiatkowska",
"Marta",
""
]
] |
2506.15066 | Jianmin Ye | Jianmin Ye, Tianyang Liu, Qi Tian, Shengchu Su, Zhe Jiang, Xi Wang | ChatModel: Automating Reference Model Design and Verification with LLMs | null | null | null | null | cs.AR cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As the complexity of integrated circuit designs continues to escalate, the functional verification becomes increasingly challenging. Reference models, critical for accelerating the verification process, are themselves becoming more intricate and time-consuming to develop. Despite the promise shown by large language models (LLMs) in code programming, effectively generating complex reference models remains a significant hurdle. To address these challenges, we introduce ChatModel, the first LLM-aided agile reference model generation and verification platform. ChatModel streamlines the transition from design specifications to fully functional reference models by integrating design standardization and hierarchical agile modeling. Employing a building-block generation strategy, it not only enhances the design capabilities of LLMs for reference models but also significantly boosts verification efficiency. We evaluated ChatModel on 300 designs of varying complexity, demonstrating substantial improvements in both efficiency and quality of reference model generation. ChatModel achieved a peak performance improvement of 55.02% compared to alternative methods, with notable enhancements in generation stability, and delivered a 9.18x increase in its capacity to produce reference model designs. Furthermore, it accelerated the iterative process of reference model design and validation by an average of 5.90x compared to traditional approaches. These results highlight the potential of ChatModel to significantly advance the automation of reference model generation and validation. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 02:15:02 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 07:31:34 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Ye",
"Jianmin",
""
],
[
"Liu",
"Tianyang",
""
],
[
"Tian",
"Qi",
""
],
[
"Su",
"Shengchu",
""
],
[
"Jiang",
"Zhe",
""
],
[
"Wang",
"Xi",
""
]
] |
2506.15196 | Xianliang Yang | Xianliang Yang, Ling Zhang, Haolong Qian, Lei Song, Jiang Bian | HeurAgenix: Leveraging LLMs for Solving Complex Combinatorial Optimization Challenges | 27 pages,9 figures | null | null | null | cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Heuristic algorithms play a vital role in solving combinatorial optimization (CO) problems, yet traditional designs depend heavily on manual expertise and struggle to generalize across diverse instances. We introduce \textbf{HeurAgenix}, a two-stage hyper-heuristic framework powered by large language models (LLMs) that first evolves heuristics and then selects among them automatically. In the heuristic evolution phase, HeurAgenix leverages an LLM to compare seed heuristic solutions with higher-quality solutions and extract reusable evolution strategies. During problem solving, it dynamically picks the most promising heuristic for each problem state, guided by the LLM's perception ability. For flexibility, this selector can be either a state-of-the-art LLM or a fine-tuned lightweight model with lower inference cost. To mitigate the scarcity of reliable supervision caused by CO complexity, we fine-tune the lightweight heuristic selector with a dual-reward mechanism that jointly exploits singals from selection preferences and state perception, enabling robust selection under noisy annotations. Extensive experiments on canonical benchmarks show that HeurAgenix not only outperforms existing LLM-based hyper-heuristics but also matches or exceeds specialized solvers. Code is available at https://github.com/microsoft/HeurAgenix. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 07:20:01 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 14:48:12 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Yang",
"Xianliang",
""
],
[
"Zhang",
"Ling",
""
],
[
"Qian",
"Haolong",
""
],
[
"Song",
"Lei",
""
],
[
"Bian",
"Jiang",
""
]
] |
2506.15201 | Jiayin Xu | Xuelin Shen, Jiayin Xu, Kangsheng Yin, Wenhan Yang | Privacy-Shielded Image Compression: Defending Against Exploitation from Vision-Language Pretrained Models | 11 pages, 6 figures, publised to ICML 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | The improved semantic understanding of vision-language pretrained (VLP) models has made it increasingly difficult to protect publicly posted images from being exploited by search engines and other similar tools. In this context, this paper seeks to protect users' privacy by implementing defenses at the image compression stage to prevent exploitation. Specifically, we propose a flexible coding method, termed Privacy-Shielded Image Compression (PSIC), that can produce bitstreams with multiple decoding options. By default, the bitstream is decoded to preserve satisfactory perceptual quality while preventing interpretation by VLP models. Our method also retains the original image compression functionality. With a customizable input condition, the proposed scheme can reconstruct the image that preserves its full semantic information. A Conditional Latent Trigger Generation (CLTG) module is proposed to produce bias information based on customizable conditions to guide the decoding process into different reconstructed versions, and an Uncertainty-Aware Encryption-Oriented (UAEO) optimization function is designed to leverage the soft labels inferred from the target VLP model's uncertainty on the training data. This paper further incorporates an adaptive multi-objective optimization strategy to obtain improved encrypting performance and perceptual quality simultaneously within a unified training process. The proposed scheme is plug-and-play and can be seamlessly integrated into most existing Learned Image Compression (LIC) models. Extensive experiments across multiple downstream tasks have demonstrated the effectiveness of our design. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 07:29:40 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 01:52:22 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Shen",
"Xuelin",
""
],
[
"Xu",
"Jiayin",
""
],
[
"Yin",
"Kangsheng",
""
],
[
"Yang",
"Wenhan",
""
]
] |
2506.15242 | Qingsong Yan | Qingsong Yan, Qiang Wang, Kaiyong Zhao, Jie Chen, Bo Li, Xiaowen Chu, Fei Deng | RA-NeRF: Robust Neural Radiance Field Reconstruction with Accurate Camera Pose Estimation under Complex Trajectories | IROS 2025 | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Neural Radiance Fields (NeRF) and 3D Gaussian Splatting (3DGS) have emerged as powerful tools for 3D reconstruction and SLAM tasks. However, their performance depends heavily on accurate camera pose priors. Existing approaches attempt to address this issue by introducing external constraints but fall short of achieving satisfactory accuracy, particularly when camera trajectories are complex. In this paper, we propose a novel method, RA-NeRF, capable of predicting highly accurate camera poses even with complex camera trajectories. Following the incremental pipeline, RA-NeRF reconstructs the scene using NeRF with photometric consistency and incorporates flow-driven pose regulation to enhance robustness during initialization and localization. Additionally, RA-NeRF employs an implicit pose filter to capture the camera movement pattern and eliminate the noise for pose estimation. To validate our method, we conduct extensive experiments on the Tanks\&Temple dataset for standard evaluation, as well as the NeRFBuster dataset, which presents challenging camera pose trajectories. On both datasets, RA-NeRF achieves state-of-the-art results in both camera pose estimation and visual quality, demonstrating its effectiveness and robustness in scene reconstruction under complex pose trajectories. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 08:21:19 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 16:47:52 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Yan",
"Qingsong",
""
],
[
"Wang",
"Qiang",
""
],
[
"Zhao",
"Kaiyong",
""
],
[
"Chen",
"Jie",
""
],
[
"Li",
"Bo",
""
],
[
"Chu",
"Xiaowen",
""
],
[
"Deng",
"Fei",
""
]
] |
2506.15297 | Chen Chen | C. Chen (1 and 2), Y. F. Niu (1 and 2), R. Xu (3), Y. Tian (3) ((1) Frontiers Science Center for Rare isotope, Lanzhou University, Lanzhou, China, (2) School of Nuclear Science and Technology, Lanzhou University, Lanzhou, China, (3) China Nuclear Data Center, China Institute of Atomic Energy, Beijing, China) | Photoabsorption Cross Sections studied within the axially deformed Relativistic Quasiparticle Finite Amplitude Framework | null | null | null | null | nucl-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Photoabsorption cross sections for 235 stable nuclei, ranging from $^{40}$Ca to $^{209}$Bi, were investigated by the quasiparticle finite amplitude method (QFAM) based on the axially deformed relativistic Hartree-Bogoliubov (RHB) approach using relativistic point-coupling interaction DD-PC1, with extensions to odd-A nuclei. GDR parameters based on the standard Lorentzian (SLO) model were extracted from QFAM results and compared with those from experimental data recommended by IAEA. Good agreement was achieved for giant dipole resonance (GDR) peak energies, while resonance widths were underestimated and hence peak cross sections were overestimated due to the lack of higher-order many-body correlations. These discrepancies were much improved in deformed nuclei. The effects of deformation on photoabsorption cross sections were examined systematically. The comparison of photoabsorption cross sections among QFAM results and discrepant experimental data revealed the potential of QFAM calculations in the evaluation of photonuclear data. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 09:28:13 GMT"
},
{
"version": "v2",
"created": "Sun, 22 Jun 2025 11:23:53 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Chen",
"C.",
"",
"1 and 2"
],
[
"Niu",
"Y. F.",
"",
"1 and 2"
],
[
"Xu",
"R.",
"",
"China Nuclear Data Center, China Institute of Atomic Energy, Beijing, China"
],
[
"Tian",
"Y.",
"",
"China Nuclear Data Center, China Institute of Atomic Energy, Beijing, China"
]
] |
2506.15496 | Maciej Kierkla | Maciej Kierkla, Nicklas Ramberg, Philipp Schicho, Daniel Schmitt | Theoretical uncertainties for primordial black holes from cosmological phase transitions | 11 pages, 1 figure | null | null | SISSA 07/2025/FISI | hep-ph astro-ph.CO | http://creativecommons.org/licenses/by/4.0/ | Strongly supercooled first-order phase transitions have been proposed as a primordial black hole (PBH) production mechanism. While previous works rely on simplified models with limited thermodynamic precision, we stress that reliable theoretical PBH predictions require precise nucleation dynamics within realistic, classically conformal extensions of the Standard Model. By employing high-temperature dimensional reduction and computing the one-loop fluctuation determinants, we provide a state-of-the-art thermodynamic analysis and estimate the corresponding PBH abundance for classically conformal gauge-Higgs theories. Accounting for constraints from successful percolation and QCD chiral symmetry breaking, the parameter space where PBHs are viable dark matter candidates is severely limited. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 14:34:48 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Kierkla",
"Maciej",
""
],
[
"Ramberg",
"Nicklas",
""
],
[
"Schicho",
"Philipp",
""
],
[
"Schmitt",
"Daniel",
""
]
] |
2506.15525 | Isabella Pu | Isabella Pu, Prerna Ravi, Linh Dieu Dinh, Chelsea Joe, Caitlin Ogoe, Zixuan Li, Cynthia Breazeal, Anastasia K. Ostrowski | "How can we learn and use AI at the same time?": Participatory Design of GenAI with High School Students | Copyright protected by ACM, 17 pages, 5 figures, 2 tables, in proceedings of 24th annual ACM Interaction Design and Children Conference (IDC 2025) | null | 10.1145/3713043.3727057 | null | cs.HC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | As generative AI (GenAI) emerges as a transformative force, clear understanding of high school students' perspectives is essential for GenAI's meaningful integration in high school environments. In this work, we draw insights from a participatory design workshop where we engaged 17 high school students -- a group rarely involved in prior research in this area -- through the design of novel GenAI tools and school policies addressing their key concerns. Students identified challenges and developed solutions outlining their ideal features in GenAI tools, appropriate school use, and regulations. These centered around the problem spaces of combating bias & misinformation, tackling crime & plagiarism, preventing over-reliance on AI, and handling false accusations of academic dishonesty. Building on our participants' underrepresented perspectives, we propose new guidelines targeted at educational technology designers for development of GenAI technologies in high schools. We also argue for further incorporation of student voices in development of AI policies in their schools. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 14:58:50 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 12:16:23 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Pu",
"Isabella",
""
],
[
"Ravi",
"Prerna",
""
],
[
"Dinh",
"Linh Dieu",
""
],
[
"Joe",
"Chelsea",
""
],
[
"Ogoe",
"Caitlin",
""
],
[
"Li",
"Zixuan",
""
],
[
"Breazeal",
"Cynthia",
""
],
[
"Ostrowski",
"Anastasia K.",
""
]
] |
2506.15546 | Giada Casali | G. Casali, J. Montalb\`an, A. Miglio, L. Casagrande, L. Magrini, C. Chiappini, A. Bragaglia, M. Matteuzzi, K. Brogaard, A. Stokholm, V. Grisoni, M. Tailo, E. Willett | Tracing the Milky Way: Calibrating chemical ages with high-precision Kepler data | 21 pages, 19 figures, accepted for publication in MNRAS | null | null | null | astro-ph.GA astro-ph.SR | http://creativecommons.org/licenses/by/4.0/ | Chemical clocks offer a powerful tool for estimating stellar ages from spectroscopic surveys. We present a new detailed spectroscopic analysis of 68 Kepler red giant stars to provide a suite of high-precision abundances along with asteroseismic ages with better than 10 percent precision from individual mode frequencies. We obtained several chemical clocks as ratios between s-process elements (Y, Zr, Ba, La, Ce) and alpha-elements (Mg, Ca, Si, Al, Ti). Our data show that [Ce/Mg] and [Zr/Ti] display a remarkably tight correlation with stellar ages, with abundance dispersions of 0.08 and 0.01 dex respectively and below 3 Gyr in ages, across the entire Galactic chronochemical history. While improving the precision floor of spectroscopic surveys is critical for broadening the scope and applicability of chemical clocks, the intrinsic accuracy of our relations -- enabled by high-resolution chemical abundances and stellar ages in our sample -- allows us to draw meaningful conclusions about age trends across stellar populations. By applying our relations to the APOGEE and Gaia-ESO surveys, we are able to differentiate the low- and high-alpha sequences in age, recover the age-metallicity relation, observe the disc flaring of the Milky Way, and identify a population of old metal-rich stars. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 15:18:40 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 12:22:12 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Casali",
"G.",
""
],
[
"Montalbàn",
"J.",
""
],
[
"Miglio",
"A.",
""
],
[
"Casagrande",
"L.",
""
],
[
"Magrini",
"L.",
""
],
[
"Chiappini",
"C.",
""
],
[
"Bragaglia",
"A.",
""
],
[
"Matteuzzi",
"M.",
""
],
[
"Brogaard",
"K.",
""
],
[
"Stokholm",
"A.",
""
],
[
"Grisoni",
"V.",
""
],
[
"Tailo",
"M.",
""
],
[
"Willett",
"E.",
""
]
] |
2506.15742 | Axel Sauer | Black Forest Labs, Stephen Batifol, Andreas Blattmann, Frederic Boesel, Saksham Consul, Cyril Diagne, Tim Dockhorn, Jack English, Zion English, Patrick Esser, Sumith Kulal, Kyle Lacey, Yam Levi, Cheng Li, Dominik Lorenz, Jonas M\"uller, Dustin Podell, Robin Rombach, Harry Saini, Axel Sauer, Luke Smith | FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space | null | null | null | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present evaluation results for FLUX.1 Kontext, a generative flow matching model that unifies image generation and editing. The model generates novel output views by incorporating semantic context from text and image inputs. Using a simple sequence concatenation approach, FLUX.1 Kontext handles both local editing and generative in-context tasks within a single unified architecture. Compared to current editing models that exhibit degradation in character consistency and stability across multiple turns, we observe that FLUX.1 Kontext improved preservation of objects and characters, leading to greater robustness in iterative workflows. The model achieves competitive performance with current state-of-the-art systems while delivering significantly faster generation times, enabling interactive applications and rapid prototyping workflows. To validate these improvements, we introduce KontextBench, a comprehensive benchmark with 1026 image-prompt pairs covering five task categories: local editing, global editing, character reference, style reference and text editing. Detailed evaluations show the superior performance of FLUX.1 Kontext in terms of both single-turn quality and multi-turn consistency, setting new standards for unified image processing models. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 20:18:23 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 05:31:03 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Labs",
"Black Forest",
""
],
[
"Batifol",
"Stephen",
""
],
[
"Blattmann",
"Andreas",
""
],
[
"Boesel",
"Frederic",
""
],
[
"Consul",
"Saksham",
""
],
[
"Diagne",
"Cyril",
""
],
[
"Dockhorn",
"Tim",
""
],
[
"English",
"Jack",
""
],
[
"English",
"Zion",
""
],
[
"Esser",
"Patrick",
""
],
[
"Kulal",
"Sumith",
""
],
[
"Lacey",
"Kyle",
""
],
[
"Levi",
"Yam",
""
],
[
"Li",
"Cheng",
""
],
[
"Lorenz",
"Dominik",
""
],
[
"Müller",
"Jonas",
""
],
[
"Podell",
"Dustin",
""
],
[
"Rombach",
"Robin",
""
],
[
"Saini",
"Harry",
""
],
[
"Sauer",
"Axel",
""
],
[
"Smith",
"Luke",
""
]
] |
2506.15746 | Kevin Xu | Kevin Xu, Risto Miikkulainen | Neural Cellular Automata for ARC-AGI | 8 pages, 5 figures | null | null | null | cs.NE cs.AI | http://creativecommons.org/licenses/by/4.0/ | Cellular automata and their differentiable counterparts, Neural Cellular Automata (NCA), are highly expressive and capable of surprisingly complex behaviors. This paper explores how NCAs perform when applied to tasks requiring precise transformations and few-shot generalization, using the Abstraction and Reasoning Corpus for Artificial General Intelligence (ARC-AGI) as a domain that challenges their capabilities in ways not previously explored. Specifically, this paper uses gradient-based training to learn iterative update rules that transform input grids into their outputs from the training examples and apply them to the test inputs. Results suggest that gradient-trained NCA models are a promising and efficient approach to a range of abstract grid-based tasks from ARC. Along with discussing the impacts of various design modifications and training constraints, this work examines the behavior and properties of NCAs applied to ARC to give insights for broader applications of self-organizing systems. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 03:47:31 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Xu",
"Kevin",
""
],
[
"Miikkulainen",
"Risto",
""
]
] |
2506.15781 | Masaki Kiyuna | Masaki Kiyuna | Super-Eddington Growth Ceiling: Analytic Constraints on the Rapid Growth of Light-Seed Black Holes in Massive Clumps | 17 pages, 7 figures, submitted to MNRAS | null | null | null | astro-ph.GA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The existence of $\sim10^{7-8}{\rm M}_\odot$ supermassive black holes at $z\gtrsim 8$ challenges conventional growth channels. One attractive possibility is that light seeds ($M_{\rm BH}\lesssim10^{3}{\rm M}_\odot$) undergo short, super-Eddington episodes when they cross, and are captured by, dense massive gas clumps. We revisit this ``BH-clump-capture'' model using analytic arguments supported by toy-model simulations that follow Bondi-scale inflow, radiative feedback, gas dynamical friction and the recently discovered forward acceleration effect caused by the ionised bubble. For substantial growth the black hole must remain trapped for many dynamical times, which imposes three simultaneous constraints. The clump must be heavier than the black hole (mass doubling condition); its cooling time must exceed the super-Eddington growth time (lifetime condition); and dynamical friction must dominate shell acceleration (BH-trapping condition). These requirements confine viable clumps to a narrow density-temperature region, $n_{\rm H}\simeq10^{7-8}{\rm cm}^{-3}$ and $T\simeq(2-6)\times10^{3}{\rm K}$, for a $10^{3}{\rm M}_\odot$ seed. Even inside this sweet spot a $10^{3}{\rm M}_\odot$ seed grow up at most $4\times10^{3}{\rm M}_\odot$; the maximum growth ratio $M_{\rm BH, fin}/M_{\rm BH}$ falls approximately as $M_{\rm BH}^{-0.4}$ and is negligible once $M_{\rm BH}\gtrsim10^{4}{\rm M}_\odot$. The forward-acceleration effect is essential, expelling the black hole whenever photon trapping fails. We conclude that BH-clump-capture model, and potentially broad super-Eddington models, cannot produce the $>10^{4}{\rm M}_\odot$ seeds required for subsequent Eddington-limited growth, suggesting that alternative pathways, such as heavy seed formation, remain necessary. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 18:00:05 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 01:30:33 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Kiyuna",
"Masaki",
""
]
] |
2506.15908 | Ulas Bagci | Elif Keles, Merve Yazol, Gorkem Durak, Ziliang Hong, Halil Ertugrul Aktas, Zheyuan Zhang, Linkai Peng, Onkar Susladkar, Necati Guzelyel, Oznur Leman Boyunaga, Cemal Yazici, Mark Lowe, Aliye Uc, Ulas Bagci | Pediatric Pancreas Segmentation from MRI Scans with Deep Learning | Code and MRI data available for public | null | 10.1016/j.pan.2025.06.006 | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Objective: Our study aimed to evaluate and validate PanSegNet, a deep learning (DL) algorithm for pediatric pancreas segmentation on MRI in children with acute pancreatitis (AP), chronic pancreatitis (CP), and healthy controls. Methods: With IRB approval, we retrospectively collected 84 MRI scans (1.5T/3T Siemens Aera/Verio) from children aged 2-19 years at Gazi University (2015-2024). The dataset includes healthy children as well as patients diagnosed with AP or CP based on clinical criteria. Pediatric and general radiologists manually segmented the pancreas, then confirmed by a senior pediatric radiologist. PanSegNet-generated segmentations were assessed using Dice Similarity Coefficient (DSC) and 95th percentile Hausdorff distance (HD95). Cohen's kappa measured observer agreement. Results: Pancreas MRI T2W scans were obtained from 42 children with AP/CP (mean age: 11.73 +/- 3.9 years) and 42 healthy children (mean age: 11.19 +/- 4.88 years). PanSegNet achieved DSC scores of 88% (controls), 81% (AP), and 80% (CP), with HD95 values of 3.98 mm (controls), 9.85 mm (AP), and 15.67 mm (CP). Inter-observer kappa was 0.86 (controls), 0.82 (pancreatitis), and intra-observer agreement reached 0.88 and 0.81. Strong agreement was observed between automated and manual volumes (R^2 = 0.85 in controls, 0.77 in diseased), demonstrating clinical reliability. Conclusion: PanSegNet represents the first validated deep learning solution for pancreatic MRI segmentation, achieving expert-level performance across healthy and diseased states. This tool, algorithm, along with our annotated dataset, are freely available on GitHub and OSF, advancing accessible, radiation-free pediatric pancreatic imaging and fostering collaborative research in this underserved domain. | [
{
"version": "v1",
"created": "Wed, 18 Jun 2025 22:59:30 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Keles",
"Elif",
""
],
[
"Yazol",
"Merve",
""
],
[
"Durak",
"Gorkem",
""
],
[
"Hong",
"Ziliang",
""
],
[
"Aktas",
"Halil Ertugrul",
""
],
[
"Zhang",
"Zheyuan",
""
],
[
"Peng",
"Linkai",
""
],
[
"Susladkar",
"Onkar",
""
],
[
"Guzelyel",
"Necati",
""
],
[
"Boyunaga",
"Oznur Leman",
""
],
[
"Yazici",
"Cemal",
""
],
[
"Lowe",
"Mark",
""
],
[
"Uc",
"Aliye",
""
],
[
"Bagci",
"Ulas",
""
]
] |
2506.15961 | Youshan Miao | Yunchi Lu, Youshan Miao, Cheng Tan, Peng Huang, Yi Zhu, Xian Zhang, Fan Yang | TrainVerify: Equivalence-Based Verification for Distributed LLM Training | null | null | null | null | cs.DC cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Training large language models (LLMs) at scale requires parallel execution across thousands of devices, incurring enormous computational costs. Yet, these costly distributed trainings are rarely verified, leaving them prone to silent errors and potentially wasting millions of GPU hours. We introduce TrainVerify, a system for verifiable distributed training of LLMs. Given a deep learning model's logical specification as the ground truth, TrainVerify formally verifies that a distributed parallel execution plan is mathematically equivalent to it. Direct verification is notoriously difficult due to the sheer scale of LLMs which often involves billions of variables and highly intricate computation graphs. Therefore, TrainVerify introduces shape-reduction techniques and a stage-wise parallel verification algorithm that significantly reduces complexity while preserving formal correctness. TrainVerify scales to frontier LLMs, including the successful verification of the Llama3 (405B) and DeepSeek-V3 (671B) training plans. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 02:10:06 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 10:50:28 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Lu",
"Yunchi",
""
],
[
"Miao",
"Youshan",
""
],
[
"Tan",
"Cheng",
""
],
[
"Huang",
"Peng",
""
],
[
"Zhu",
"Yi",
""
],
[
"Zhang",
"Xian",
""
],
[
"Yang",
"Fan",
""
]
] |
2506.16058 | Yong Liu | Yong Liu, SongLi Wu, Sule Bai, Jiahao Wang, Yitong Wang, Yansong Tang | Stepping Out of Similar Semantic Space for Open-Vocabulary Segmentation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-sa/4.0/ | Open-vocabulary segmentation aims to achieve segmentation of arbitrary categories given unlimited text inputs as guidance. To achieve this, recent works have focused on developing various technical routes to exploit the potential of large-scale pre-trained vision-language models and have made significant progress on existing benchmarks. However, we find that existing test sets are limited in measuring the models' comprehension of ``open-vocabulary" concepts, as their semantic space closely resembles the training space, even with many overlapping categories. To this end, we present a new benchmark named OpenBench that differs significantly from the training semantics. It is designed to better assess the model's ability to understand and segment a wide range of real-world concepts. When testing existing methods on OpenBench, we find that their performance diverges from the conclusions drawn on existing test sets. In addition, we propose a method named OVSNet to improve the segmentation performance for diverse and open scenarios. Through elaborate fusion of heterogeneous features and cost-free expansion of the training space, OVSNet achieves state-of-the-art results on both existing datasets and our proposed OpenBench. Corresponding analysis demonstrate the soundness and effectiveness of our proposed benchmark and method. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 06:32:53 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 03:11:42 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Liu",
"Yong",
""
],
[
"Wu",
"SongLi",
""
],
[
"Bai",
"Sule",
""
],
[
"Wang",
"Jiahao",
""
],
[
"Wang",
"Yitong",
""
],
[
"Tang",
"Yansong",
""
]
] |
2506.16218 | Xinting Liao | Xinting Liao, Weiming Liu, Jiaming Qian, Pengyang Zhou, Jiahe Xu, Wenjie Wang, Chaochao Chen, Xiaolin Zheng, Tat-Seng Chua | FOCoOp: Enhancing Out-of-Distribution Robustness in Federated Prompt Learning for Vision-Language Models | Accepted by ICML25 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Federated prompt learning (FPL) for vision-language models is a powerful approach to collaboratively adapt models across distributed clients while preserving data privacy. However, existing FPL approaches suffer from a trade-off between performance and robustness, particularly in out-of-distribution (OOD) shifts, limiting their reliability in real-world scenarios. The inherent in-distribution (ID) data heterogeneity among different clients makes it more challenging to maintain this trade-off. To fill this gap, we introduce a Federated OOD-aware Context Optimization (FOCoOp) framework, which captures diverse distributions among clients using ID global prompts, local prompts, and OOD prompts. Specifically, FOCoOp leverages three sets of prompts to create both class-level and distribution-level separations, which adapt to OOD shifts through bi-level distributionally robust optimization. Additionally, FOCoOp improves the discrimination consistency among clients, i.e., calibrating global prompts, seemingly OOD prompts, and OOD prompts by semi-unbalanced optimal transport. The extensive experiments on real-world datasets demonstrate that FOCoOp effectively captures decentralized heterogeneous distributions and enhances robustness of different OOD shifts. The project is available at GitHub. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 11:16:02 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 14:51:04 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Liao",
"Xinting",
""
],
[
"Liu",
"Weiming",
""
],
[
"Qian",
"Jiaming",
""
],
[
"Zhou",
"Pengyang",
""
],
[
"Xu",
"Jiahe",
""
],
[
"Wang",
"Wenjie",
""
],
[
"Chen",
"Chaochao",
""
],
[
"Zheng",
"Xiaolin",
""
],
[
"Chua",
"Tat-Seng",
""
]
] |
2506.16245 | Wen-Bin Chang | Wen-Bin Chang | Schwinger Effect in a Twice Anisotropic Holographic Model | null | null | null | null | hep-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this work, we investigate the Schwinger effect in a twice anisotropic holographic QCD model that incorporates both spatial and magnetic anisotropies. Using the AdS/CFT correspondence, we calculate the total potential of a particle-antiparticle pair to evaluate how these anisotropies affect the holographic Schwinger effect. Our calculations reveal that the magnetic field, characterized by parameters $c_B$ and $q_3$, consistently enhances the Schwinger effect by lowering and narrowing the potential barrier. In contrast, increasing the spatial anisotropy, parameterized by $\nu$, raises and widens the barrier, thereby suppressing the process. These findings suggest the significance of treating both anisotropies concurrently for a realistic description of particle production in HIC. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 12:01:40 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 02:31:50 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Jun 2025 02:24:39 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Chang",
"Wen-Bin",
""
]
] |
2506.16297 | Zikang Wan | Heng Zhang, Zikang Wan, Danilo Vasconcellos Vargas | SycnMapV2: Robust and Adaptive Unsupervised Segmentation | null | null | null | null | cs.CV cs.AI cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Human vision excels at segmenting visual cues without the need for explicit training, and it remains remarkably robust even as noise severity increases. In contrast, existing AI algorithms struggle to maintain accuracy under similar conditions. Here, we present SyncMapV2, the first to solve unsupervised segmentation with state-of-the-art robustness. SyncMapV2 exhibits a minimal drop in mIoU, only 0.01%, under digital corruption, compared to a 23.8% drop observed in SOTA methods. This superior performance extends across various types of corruption: noise (7.3% vs. 37.7%), weather (7.5% vs. 33.8%), and blur (7.0% vs. 29.5%). Notably, SyncMapV2 accomplishes this without any robust training, supervision, or loss functions. It is based on a learning paradigm that uses self-organizing dynamical equations combined with concepts from random networks. Moreover, unlike conventional methods that require re-initialization for each new input, SyncMapV2 adapts online, mimicking the continuous adaptability of human vision. Thus, we go beyond the accurate and robust results, and present the first algorithm that can do all the above online, adapting to input rather than re-initializing. In adaptability tests, SyncMapV2 demonstrates near-zero performance degradation, which motivates and fosters a new generation of robust and adaptive intelligence in the near future. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 13:17:30 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 04:07:21 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Zhang",
"Heng",
""
],
[
"Wan",
"Zikang",
""
],
[
"Vargas",
"Danilo Vasconcellos",
""
]
] |
2506.16640 | Marcos Vin\'icius Treviso | Pavlo Vasylenko, Marcos Treviso, Andr\'e F. T. Martins | Long-Context Generalization with Sparse Attention | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Transformer-based architectures traditionally employ softmax to compute attention weights, which produces dense distributions over all tokens in a sequence. While effective in many settings, this density has been shown to be detrimental for tasks that demand precise focus on fixed-size patterns: as sequence length increases, non-informative tokens accumulate attention probability mass, leading to dispersion and representational collapse. We show in this paper that sparse attention mechanisms using $\alpha$-entmax can avoid these issues, due to their ability to assign exact zeros to irrelevant tokens. Furthermore, we introduce Adaptive-Scalable Entmax (ASEntmax), which endows $\alpha$-entmax with a learnable temperature parameter, allowing the attention distribution to interpolate between sparse (pattern-focused) and dense (softmax-like) regimes. Finally, we show that the ability to locate and generalize fixed-size patterns can be further improved through a careful design of position encodings, which impacts both dense and sparse attention methods. By integrating ASEntmax into standard transformer layers alongside proper positional encodings, we show that our models greatly outperform softmax, scalable softmax, and fixed-temperature $\alpha$-entmax baselines on long-context generalization. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 22:43:25 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 04:45:00 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Vasylenko",
"Pavlo",
""
],
[
"Treviso",
"Marcos",
""
],
[
"Martins",
"André F. T.",
""
]
] |
2506.16687 | Alexander Lange | Alexander Lange, J. Eagle, O. Kargaltsev, Lucien Kuiper, and Jeremy Hare | The Vela pulsar and its pulsar wind nebula Vela-X using 13 years of Fermi-LAT Observations | Accepted for publication in the Astrophysical Journal. 24 pages, 15 figures | null | null | null | astro-ph.HE | http://creativecommons.org/licenses/by/4.0/ | We present results of more than 13 years of Fermi-LAT data analysis for the Vela pulsar from 60 MeV to 100 GeV and its pulsar wind nebula (PWN), Vela-X, for E > 1 GeV in the off-pulse phases. We find the Vela-X PWN can be best characterized using two extended components: a large radial Gaussian accompanied by an off-set, compact radial disk, both with a similar spectral index, \Gamma \sim 2.3. The common spectral properties support a common PWN origin, but a supernova remnant component is plausible for the compact radial disk. With an updated Vela-X model, the phase resolved spectral properties of the Vela pulsar are explored through a phase-resolved analysis. The phase-resolved spectral properties of the pulsar are presented, such as the SED peak energy E$_p$, the width of the SED at its peak, d$_p$, and the asymptotic (low-energy) spectral index, $\Gamma_0$, are presented. The best-fit spectral models for each LAT pulse peak (Peak 1 and Peak 2) are extrapolated to UV energies and compared to archival, phase-resolved spectra at UV, X-ray, soft \gamma-ray and TeV energies. We also discuss the physical implications of our modeling and the data comparisons. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 01:58:26 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 17:13:39 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Lange",
"Alexander",
""
],
[
"Eagle",
"J.",
""
],
[
"Kargaltsev",
"O.",
""
],
[
"Kuiper",
"Lucien",
""
],
[
"Hare",
"Jeremy",
""
]
] |
2506.16710 | Souma Chowdhury | Aditya Bhatt, Mary Katherine Corra, Franklin Merlo, Prajit KrisshnaKumar and Souma Chowdhury | Experimental Setup and Software Pipeline to Evaluate Optimization based Autonomous Multi-Robot Search Algorithms | IDETC 2025 | null | null | null | cs.RO cs.MA | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Signal source localization has been a problem of interest in the multi-robot systems domain given its applications in search & rescue and hazard localization in various industrial and outdoor settings. A variety of multi-robot search algorithms exist that usually formulate and solve the associated autonomous motion planning problem as a heuristic model-free or belief model-based optimization process. Most of these algorithms however remains tested only in simulation, thereby losing the opportunity to generate knowledge about how such algorithms would compare/contrast in a real physical setting in terms of search performance and real-time computing performance. To address this gap, this paper presents a new lab-scale physical setup and associated open-source software pipeline to evaluate and benchmark multi-robot search algorithms. The presented physical setup innovatively uses an acoustic source (that is safe and inexpensive) and small ground robots (e-pucks) operating in a standard motion-capture environment. This setup can be easily recreated and used by most robotics researchers. The acoustic source also presents interesting uncertainty in terms of its noise-to-signal ratio, which is useful to assess sim-to-real gaps. The overall software pipeline is designed to readily interface with any multi-robot search algorithm with minimal effort and is executable in parallel asynchronous form. This pipeline includes a framework for distributed implementation of multi-robot or swarm search algorithms, integrated with a ROS (Robotics Operating System)-based software stack for motion capture supported localization. The utility of this novel setup is demonstrated by using it to evaluate two state-of-the-art multi-robot search algorithms, based on swarm optimization and batch-Bayesian Optimization (called Bayes-Swarm), as well as a random walk baseline. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 03:06:43 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 22:01:07 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Bhatt",
"Aditya",
""
],
[
"Corra",
"Mary Katherine",
""
],
[
"Merlo",
"Franklin",
""
],
[
"KrisshnaKumar",
"Prajit",
""
],
[
"Chowdhury",
"Souma",
""
]
] |
2506.16757 | Xuankai Xu | Xuankai Xu, Jiawei Li, Ruoyu Wang, Ruihong Xiong, Yiwei Wang, Xiaoqin Shen, Tao Wu | Piezoelectric-Metal Phononic Crystal Enabling GHz Tunable Ultrahigh $Q$ Quasi-BIC mode | null | null | null | null | physics.app-ph | http://creativecommons.org/licenses/by/4.0/ | The integration of GHz-frequency, high quality factor ($Q$), and electrically tunable acoustic resonators holds significant potential for advancing applications in quantum information technologies, microwave photonics, and reconfigurable RF systems. However, simultaneously achieving these three characteristics within a single, scalable platform remains a fundamental challenge. Here, we report the experimental demonstration of a GHz quasi-BIC resonator in a piezoelectric thin-film shear horizontal (SH) wave system, achieved through a structurally simple piezoelectric-metal phononic crystal (PnC) architecture on a LiNbO$_3$ thin film. This approach enables leaky Fabry-Perot coupling mode and localized trapping quasi-BIC mode. Without the need for deep etching or intricate patterning, we achieve a room-temperature quality factor of $6\times 10^4$ at ~1 GHz in ambient air, corresponding to an $f\times Q$ product of $6\times 10^{13}$ Hz at quasi-BIC mode. Furthermore, we demonstrate efficient electrical tunability via low-voltage (0.6 V) electrothermal modulation of the PnC structure, enabling a reversible transition between trapped and transmission states and yielding a high-contrast amplitude modulation of 47.75 dB. Our results establish a lithography-friendly, fabrication-tolerant platform for realizing tunable, high-$Q$ acoustic resonators at GHz frequencies, overcoming longstanding barriers in phononic device engineering. This work opens new directions for scalable on-chip phononic circuits in quantum acoustics, reconfigurable RF systems, and signal processing applications. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 05:25:16 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 01:00:28 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Xu",
"Xuankai",
""
],
[
"Li",
"Jiawei",
""
],
[
"Wang",
"Ruoyu",
""
],
[
"Xiong",
"Ruihong",
""
],
[
"Wang",
"Yiwei",
""
],
[
"Shen",
"Xiaoqin",
""
],
[
"Wu",
"Tao",
""
]
] |
2506.16859 | Sha Wang | Yang Hu, Minyu Fan, Kun Liu, Songsong Zhu, Nan Jiang, Sha Wang | RS-Coded Adaptive Dynamic Network for Reliable Long-Term Information Transmission in Disturbed Multimode Fiber | null | null | null | null | physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimode fiber (MMF), due to its large core diameter and high mode capacity, holds potential in high-speed communications. However, inherent modal dispersion causes output speckle distortion, and transmission characteristics are sensitive to environmental disturbances, limiting its reliable application. Conventional transmission matrix (TM) methods face challenges such as complex calibration and environmental sensitivity. Although current deep learning approaches demonstrate reconstruction potential, they struggle to overcome error accumulation caused by fiber mode drift and lack sufficient environmental adaptability. To address this, the present study proposes an adaptive transmission framework named Residual Reed-Solomon Dynamic Network (RRSDN), which integrates Reed-Solomon (RS) error correction coding with deep residual learning forming a closed-loop system that jointly optimizes encoding, transmission, and reconstruction, to tackle the key challenges of mode instability and error accumulation in dynamic scattering channels. Experimentally, high-fidelity real-time transmission of a 16*16 pixel video stream (H.265 compressed) with zero frame loss and 100% symbol accuracy was achieved under conditions of a 100-meter MMF with manually applied disturbances and no temperature control. This work proposes a solution for stable optical transmission in complex channels. Plus, it integrates error correction coding with neural network training, laying the foundation for adaptive optical systems in longer-distance and more complex scenarios. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 09:17:45 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 03:37:19 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Hu",
"Yang",
""
],
[
"Fan",
"Minyu",
""
],
[
"Liu",
"Kun",
""
],
[
"Zhu",
"Songsong",
""
],
[
"Jiang",
"Nan",
""
],
[
"Wang",
"Sha",
""
]
] |
2506.16986 | Yuntao Ma | Yuntao Ma, Yang Liu, Kaixian Qu, Marco Hutter | Learning Accurate Whole-body Throwing with High-frequency Residual Policy and Pullback Tube Acceleration | 8 pages, IROS 2025 | null | null | null | cs.RO | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Throwing is a fundamental skill that enables robots to manipulate objects in ways that extend beyond the reach of their arms. We present a control framework that combines learning and model-based control for prehensile whole-body throwing with legged mobile manipulators. Our framework consists of three components: a nominal tracking policy for the end-effector, a high-frequency residual policy to enhance tracking accuracy, and an optimization-based module to improve end-effector acceleration control. The proposed controller achieved the average of 0.28 m landing error when throwing at targets located 6 m away. Furthermore, in a comparative study with university students, the system achieved a velocity tracking error of 0.398 m/s and a success rate of 56.8%, hitting small targets randomly placed at distances of 3-5 m while throwing at a specified speed of 6 m/s. In contrast, humans have a success rate of only 15.2%. This work provides an early demonstration of prehensile throwing with quantified accuracy on hardware, contributing to progress in dynamic whole-body manipulation. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 13:29:59 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 14:07:21 GMT"
},
{
"version": "v3",
"created": "Tue, 24 Jun 2025 02:07:27 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Ma",
"Yuntao",
""
],
[
"Liu",
"Yang",
""
],
[
"Qu",
"Kaixian",
""
],
[
"Hutter",
"Marco",
""
]
] |
2506.17096 | Domantas Burba | Domantas Burba and Gediminas Juzeli\=unas | Two dimensional sub-wavelength topological dark state lattices | 11 pages, 8 figures | null | null | null | cond-mat.quant-gas cond-mat.other physics.atom-ph quant-ph | http://creativecommons.org/licenses/by/4.0/ | We present a general framework for engineering two-dimensional (2D) sub-wavelength topological optical lattices using spatially dependent atomic dark states in a $\Lambda$-type configuration of the atom-light coupling. By properly designing the spatial profiles of the laser fields inducing coupling between the atomic internal states, we show how to generate sub-wavelength Kronig-Penney-like geometric scalar potential accompanied by narrow and strong patches of the synthetic magnetic field localized in the same areas as the scalar potential. These sharply peaked magnetic fluxes are compensated by a smooth background magnetic field of opposite sign, resulting in zero net flux per unit cell while still enabling topologically nontrivial band structures. Specifically, for sufficiently narrow peaks, their influence is minimum, and the behavior of the system in a remaining smooth background magnetic field resembles the Landau problem, allowing for the formation of nearly flat energy bands with unit Chern numbers. Numerical analysis confirms the existence of ideal Chern bands and the robustness of the topological phases against non-adiabatic effects and losses. This makes the scheme well-suited for simulating quantum Hall systems and fractional Chern insulators in ultracold atomic gases, offering a new platform for exploring strongly correlated topological phases with high tunability. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 16:03:52 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 17:13:25 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Burba",
"Domantas",
""
],
[
"Juzeliūnas",
"Gediminas",
""
]
] |
2506.17260 | Liqun Qi | Chunfeng Cui, Liqun Qi and Yi Xu | PSD and SOS Biquadratic Polynomials | null | null | null | null | math.OC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hilbert indicated that a positive semi-definite (psd) homogeneous quartic polynomial of four variables may not be sum of squares (sos). {Since a} $2 \times 2$ biquadratic polynomial is a homogeneous quartic polynomial of four variables{, this raises a fundamental question:} Is a psd $2 \times 2$ biquadratic polynomial sos? In this paper, we present a necessary and sufficient condition for a $2 \times 2$ psd biquadratic polynomial to be sos. We show that if a $2 \times 2$ {psd} biquadratic polynomial is sos, then it is sos of tetranomials, its sos rank is at most $4$, and the coefficients of the tetranomials can be orthogonal to each other. We prove that every psd $2\times 2$ biquadratic polynomial admits an equivalent representation with two neighbor half-cross terms and one full-cross term. Furthermore, any $2 \times 2$ psd biquadratic polynomial is a sum of squares of binomials (sosb) when it contains at most one half-cross term, and a sum of squares of trinomials (sostri) when it contains exactly two neighbor half-cross terms and no full-cross term. | [
{
"version": "v1",
"created": "Mon, 09 Jun 2025 13:35:43 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 08:56:49 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Cui",
"Chunfeng",
""
],
[
"Qi",
"Liqun",
""
],
[
"Xu",
"Yi",
""
]
] |
2506.17336 | Minchan Kim | Yubeen Bae, Minchan Kim, Jaejin Lee, Sangbum Kim, Jaehyung Kim, Yejin Choi, Niloofar Mireshghallah | Privacy-Preserving LLM Interaction with Socratic Chain-of-Thought Reasoning and Homomorphically Encrypted Vector Databases | 29 pages | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Large language models (LLMs) are increasingly used as personal agents, accessing sensitive user data such as calendars, emails, and medical records. Users currently face a trade-off: They can send private records, many of which are stored in remote databases, to powerful but untrusted LLM providers, increasing their exposure risk. Alternatively, they can run less powerful models locally on trusted devices. We bridge this gap. Our Socratic Chain-of-Thought Reasoning first sends a generic, non-private user query to a powerful, untrusted LLM, which generates a Chain-of-Thought (CoT) prompt and detailed sub-queries without accessing user data. Next, we embed these sub-queries and perform encrypted sub-second semantic search using our Homomorphically Encrypted Vector Database across one million entries of a single user's private data. This represents a realistic scale of personal documents, emails, and records accumulated over years of digital activity. Finally, we feed the CoT prompt and the decrypted records to a local language model and generate the final response. On the LoCoMo long-context QA benchmark, our hybrid framework, combining GPT-4o with a local Llama-3.2-1B model, outperforms using GPT-4o alone by up to 7.1 percentage points. This demonstrates a first step toward systems where tasks are decomposed and split between untrusted strong LLMs and weak local ones, preserving user privacy. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 07:13:30 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Bae",
"Yubeen",
""
],
[
"Kim",
"Minchan",
""
],
[
"Lee",
"Jaejin",
""
],
[
"Kim",
"Sangbum",
""
],
[
"Kim",
"Jaehyung",
""
],
[
"Choi",
"Yejin",
""
],
[
"Mireshghallah",
"Niloofar",
""
]
] |
2506.17338 | Duong Bach | Duong Bach | PBFT-Backed Semantic Voting for Multi-Agent Memory Pruning | 13 pages | null | null | null | cs.DC cs.AI cs.MA | http://creativecommons.org/licenses/by-sa/4.0/ | The proliferation of multi-agent systems (MAS) in complex, dynamic environments necessitates robust and efficient mechanisms for managing shared knowledge. A critical challenge is ensuring that distributed memories remain synchronized, relevant, and free from the accumulation of outdated or inconsequential data - a process analogous to biological forgetting. This paper introduces the Co-Forgetting Protocol, a novel, comprehensive framework designed to address this challenge by enabling synchronized memory pruning in MAS. The protocol integrates three key components: (1) context-aware semantic voting, where agents utilize a lightweight DistilBERT model to assess the relevance of memory items based on their content and the current operational context; (2) multi-scale temporal decay functions, which assign diminishing importance to memories based on their age and access frequency across different time horizons; and (3) a Practical Byzantine Fault Tolerance (PBFT)-based consensus mechanism, ensuring that decisions to retain or discard memory items are agreed upon by a qualified and fault-tolerant majority of agents, even in the presence of up to f Byzantine (malicious or faulty) agents in a system of N greater than or equal to 3f+1 agents. The protocol leverages gRPC for efficient inter-agent communication and Pinecone for scalable vector embedding storage and similarity search, with SQLite managing metadata. Experimental evaluations in a simulated MAS environment with four agents demonstrate the protocol's efficacy, achieving a 52% reduction in memory footprint over 500 epochs, 88% voting accuracy in forgetting decisions against human-annotated benchmarks, a 92% PBFT consensus success rate under simulated Byzantine conditions, and an 82% cache hit rate for memory access. | [
{
"version": "v1",
"created": "Thu, 19 Jun 2025 08:28:29 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 06:44:47 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Bach",
"Duong",
""
]
] |
2506.17361 | Xufei Wang | Xufei Wang, Mingjian Zhang, Fei Ge, Jinchen Zhu, Wen Sha, Jifen Ren, Zhimeng Hou, Shouguo Zheng, ling Zheng, Shizhuang Weng | Efficient Feedback Gate Network for Hyperspectral Image Super-Resolution | 20 pages,17 figures | null | null | null | eess.IV cs.CV cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Even without auxiliary images, single hyperspectral image super-resolution (SHSR) methods can be designed to improve the spatial resolution of hyperspectral images. However, failing to explore coherence thoroughly along bands and spatial-spectral information leads to the limited performance of the SHSR. In this study, we propose a novel group-based SHSR method termed the efficient feedback gate network, which uses various feedbacks and gate operations involving large kernel convolutions and spectral interactions. In particular, by providing different guidance for neighboring groups, we can learn rich band information and hierarchical hyperspectral spatial information using channel shuffling and dilatation convolution in shuffled and progressive dilated fusion module(SPDFM). Moreover, we develop a wide-bound perception gate block and a spectrum enhancement gate block to construct the spatial-spectral reinforcement gate module (SSRGM) and obtain highly representative spatial-spectral features efficiently. Additionally, we apply a three-dimensional SSRGM to enhance holistic information and coherence for hyperspectral data. The experimental results on three hyperspectral datasets demonstrate the superior performance of the proposed network over the state-of-the-art methods in terms of spectral fidelity and spatial content reconstruction. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 10:06:48 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Wang",
"Xufei",
""
],
[
"Zhang",
"Mingjian",
""
],
[
"Ge",
"Fei",
""
],
[
"Zhu",
"Jinchen",
""
],
[
"Sha",
"Wen",
""
],
[
"Ren",
"Jifen",
""
],
[
"Hou",
"Zhimeng",
""
],
[
"Zheng",
"Shouguo",
""
],
[
"Zheng",
"ling",
""
],
[
"Weng",
"Shizhuang",
""
]
] |
2506.17364 | Alvaro Becerra | Alvaro Becerra, Roberto Daza, Ruth Cobos, Aythami Morales, Mutlu Cukurova, Julian Fierrez | AI-based Multimodal Biometrics for Detecting Smartphone Distractions: Application to Online Learning | Accepted in EC-TEL25: 20th European Conference on Technology Enhanced Learning, Newcastle and Durham, UK, 15-19 September 2025 | null | null | null | cs.CY cs.AI cs.CV cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This work investigates the use of multimodal biometrics to detect distractions caused by smartphone use during tasks that require sustained attention, with a focus on computer-based online learning. Although the methods are applicable to various domains, such as autonomous driving, we concentrate on the challenges learners face in maintaining engagement amid internal (e.g., motivation), system-related (e.g., course design) and contextual (e.g., smartphone use) factors. Traditional learning platforms often lack detailed behavioral data, but Multimodal Learning Analytics (MMLA) and biosensors provide new insights into learner attention. We propose an AI-based approach that leverages physiological signals and head pose data to detect phone use. Our results show that single biometric signals, such as brain waves or heart rate, offer limited accuracy, while head pose alone achieves 87%. A multimodal model combining all signals reaches 91% accuracy, highlighting the benefits of integration. We conclude by discussing the implications and limitations of deploying these models for real-time support in online learning environments. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 11:37:19 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 13:38:12 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Becerra",
"Alvaro",
""
],
[
"Daza",
"Roberto",
""
],
[
"Cobos",
"Ruth",
""
],
[
"Morales",
"Aythami",
""
],
[
"Cukurova",
"Mutlu",
""
],
[
"Fierrez",
"Julian",
""
]
] |
2506.17385 | Dorian Amaral | Dorian W. P. Amaral and Tim M. Fuchs and Hendrik Ulbricht and Christopher D. Tunnell | The MORRIS Experiment: Magnetic Levitation as a New Probe of Non-Newtonian Gravity | 11 pages, 6 figures, 1 table | null | null | null | hep-ph gr-qc hep-ex quant-ph | http://creativecommons.org/licenses/by/4.0/ | We present MORRIS (Magnetic Oscillatory Resonator for Rare-Interaction Studies) and propose the first tabletop search for non-Newtonian gravity due to a Yukawa-like fifth force using a magnetically levitated particle. Our experiment comprises a levitated sub-millimeter magnet in a superconducting trap that is driven by a time-periodic source. Featuring short-, medium-, and long-term stages, MORRIS will admit increasing sensitivities to the force coupling strength $\alpha$, optimally probing screening lengths of $\lambda \sim 1\,\mathrm{mm}$. Our short-term setup provides a proof-of-principle study, with our medium- and long-term stages respectively constraining $\alpha \lesssim 10^{-4}$ and $\alpha \lesssim 10^{-5}$, leading over existing bounds. Our projections are readily recastable to concrete models predicting the existence of fifth forces, and our statistical analysis is generally applicable to well-characterized sinusoidal driving forces. By leveraging ultralow dissipation and heavy test masses, MORRIS opens a new window onto tests of small-scale gravity and searches for physics beyond the Standard Model. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 18:00:00 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Amaral",
"Dorian W. P.",
""
],
[
"Fuchs",
"Tim M.",
""
],
[
"Ulbricht",
"Hendrik",
""
],
[
"Tunnell",
"Christopher D.",
""
]
] |
2506.17412 | Zijun Sun | Zijun Sun, Solveig Thrun, Michael Kampffmeyer | VMRA-MaR: An Asymmetry-Aware Temporal Framework for Longitudinal Breast Cancer Risk Prediction | MICCAI 2025, Provisional Accept | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Breast cancer remains a leading cause of mortality worldwide and is typically detected via screening programs where healthy people are invited in regular intervals. Automated risk prediction approaches have the potential to improve this process by facilitating dynamically screening of high-risk groups. While most models focus solely on the most recent screening, there is growing interest in exploiting temporal information to capture evolving trends in breast tissue, as inspired by clinical practice. Early methods typically relied on two time steps, and although recent efforts have extended this to multiple time steps using Transformer architectures, challenges remain in fully harnessing the rich temporal dynamics inherent in longitudinal imaging data. In this work, we propose to instead leverage Vision Mamba RNN (VMRNN) with a state-space model (SSM) and LSTM-like memory mechanisms to effectively capture nuanced trends in breast tissue evolution. To further enhance our approach, we incorporate an asymmetry module that utilizes a Spatial Asymmetry Detector (SAD) and Longitudinal Asymmetry Tracker (LAT) to identify clinically relevant bilateral differences. This integrated framework demonstrates notable improvements in predicting cancer onset, especially for the more challenging high-density breast cases and achieves superior performance at extended time points (years four and five), highlighting its potential to advance early breast cancer recognition and enable more personalized screening strategies. Our code is available at https://github.com/Mortal-Suen/VMRA-MaR.git. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 18:17:19 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Sun",
"Zijun",
""
],
[
"Thrun",
"Solveig",
""
],
[
"Kampffmeyer",
"Michael",
""
]
] |
2506.17425 | Minmin Yang | Minmin Yang, Huantao Ren, Senem Velipasalar | Trans${^2}$-CBCT: A Dual-Transformer Framework for Sparse-View CBCT Reconstruction | null | null | null | null | eess.IV cs.AI cs.CV | http://creativecommons.org/licenses/by/4.0/ | Cone-beam computed tomography (CBCT) using only a few X-ray projection views enables faster scans with lower radiation dose, but the resulting severe under-sampling causes strong artifacts and poor spatial coverage. We address these challenges in a unified framework. First, we replace conventional UNet/ResNet encoders with TransUNet, a hybrid CNN-Transformer model. Convolutional layers capture local details, while self-attention layers enhance global context. We adapt TransUNet to CBCT by combining multi-scale features, querying view-specific features per 3D point, and adding a lightweight attenuation-prediction head. This yields Trans-CBCT, which surpasses prior baselines by 1.17 dB PSNR and 0.0163 SSIM on the LUNA16 dataset with six views. Second, we introduce a neighbor-aware Point Transformer to enforce volumetric coherence. This module uses 3D positional encoding and attention over k-nearest neighbors to improve spatial consistency. The resulting model, Trans$^2$-CBCT, provides an additional gain of 0.63 dB PSNR and 0.0117 SSIM. Experiments on LUNA16 and ToothFairy show consistent gains from six to ten views, validating the effectiveness of combining CNN-Transformer features with point-based geometry reasoning for sparse-view CBCT reconstruction. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 18:45:12 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Yang",
"Minmin",
""
],
[
"Ren",
"Huantao",
""
],
[
"Velipasalar",
"Senem",
""
]
] |
2506.17441 | Florian Kogelbauer | Florian Kogelbauer and Ilya Karlin | On the Relation of Exact Hydrodynamics to the Chapman-Enskog Series | null | null | null | null | math-ph math.AP math.MP | http://creativecommons.org/licenses/by/4.0/ | We demonstrate that the Chapman-Enskog series is locally equivalent to the exact spectral closure defined on slow kinetic eigenmodes in the limit of vanishing Knudsen number. We further show that the Chapman-Enskog series diverges everywhere expect at the global equilibrium for an explicit example, while the exact spectrally closed hydrodynamics are defined globally for any Knudsen number. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 19:21:03 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Kogelbauer",
"Florian",
""
],
[
"Karlin",
"Ilya",
""
]
] |
2506.17459 | Siyu Liang | Siyu Liang, Gina-Anne Levow | Breaking the Transcription Bottleneck: Fine-tuning ASR Models for Extremely Low-Resource Fieldwork Languages | null | null | null | null | cs.CL cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Automatic Speech Recognition (ASR) has reached impressive accuracy for high-resource languages, yet its utility in linguistic fieldwork remains limited. Recordings collected in fieldwork contexts present unique challenges, including spontaneous speech, environmental noise, and severely constrained datasets from under-documented languages. In this paper, we benchmark the performance of two fine-tuned multilingual ASR models, MMS and XLS-R, on five typologically diverse low-resource languages with control of training data duration. Our findings show that MMS is best suited when extremely small amounts of training data are available, whereas XLS-R shows parity performance once training data exceed one hour. We provide linguistically grounded analysis for further provide insights towards practical guidelines for field linguists, highlighting reproducible ASR adaptation approaches to mitigate the transcription bottleneck in language documentation. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 19:59:49 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Liang",
"Siyu",
""
],
[
"Levow",
"Gina-Anne",
""
]
] |
2506.17469 | Thomas Plante St-Cyr | Thomas Plante St-Cyr, Fran\c{c}ois Duhaime, Jean-S\'ebastien Dub\'e, Simon Grenier | Dataset of soil images with corresponding particle size distributions for photogranulometry | 8 pages, 10 figures, conference | null | null | null | cs.CV eess.IV | http://creativecommons.org/licenses/by/4.0/ | Traditional particle size distribution (PSD) analyses create significant downtime and are expensive in labor and maintenance. These drawbacks could be alleviated using optical grain size analysis integrated into routine geotechnical laboratory workflow. This paper presents a high-resolution dataset of 12,714 images of 321 different soil samples collected in the Montreal, Quebec region, alongside their PSD analysis. It is designed to provide a robust starting point for training convolutional neural networks (CNN) in geotechnical applications. Soil samples were photographed in a standardized top-view position with a resolution of 45 MP and a minimum scale of 39.4 micrometers per pixel, both in their moist and dry states. A custom test bench employing 13x9 inch white aluminum trays, on which the samples are spread in a thin layer, was used. For samples exceeding a size limit, a coning and quartering method was employed for mass reduction. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 20:19:45 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 03:07:00 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"St-Cyr",
"Thomas Plante",
""
],
[
"Duhaime",
"François",
""
],
[
"Dubé",
"Jean-Sébastien",
""
],
[
"Grenier",
"Simon",
""
]
] |
2506.17513 | Rudra Bedekar | Rudra Y. Bedekar | Public Perceptions of Autonomous Vehicles: A Survey of Pedestrians and Cyclists in Pittsburgh | null | null | null | PIT DOT AV Survey 2021 | cs.CY cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study investigates how autonomous vehicle(AV) technology is perceived by pedestrians and bicyclists in Pittsburgh. Using survey data from over 1200 respondents, the research explores the interplay between demographics, AV interactions, infrastructural readiness, safety perceptions, and trust. Findings highlight demographic divides, infrastructure gaps, and the crucial role of communication and education in AV adoption. | [
{
"version": "v1",
"created": "Fri, 20 Jun 2025 23:27:06 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Bedekar",
"Rudra Y.",
""
]
] |
2506.17522 | Steffen Hallmann | RNO-G Collaboration: S. Agarwal, J. A. Aguilar, N. Alden, S. Ali, P. Allison, M. Betts, D. Besson, A. Bishop, O. Botner, S. Bouma, S. Buitink, R. Camphyn, J. Chan, S. Chiche, B. A. Clark, A. Coleman, K. Couberly, S. de Kockere, K. D. de Vries, C. Deaconu, P. Giri, C. Glaser, T. Gl\"usenkamp, H. Gui, A. Hallgren, S. Hallmann, J. C. Hanson, K. Helbing, B. Hendricks, J. Henrichs, N. Heyer, C. Hornhuber, E. Huesca Santiago, K. Hughes, A. Jaitly, T. Karg, A. Karle, J. L. Kelley, J. Kimo, C. Kopper, M. Korntheuer, M. Kowalski, I. Kravchenko, R. Krebs, M. Kugelmeier, R. Lahmann, C.-H. Liu, M. J. Marsee, Z. S. Meyers, K. Mulrey, M. Muzio, A. Nelles, A. Novikov, A. Nozdrina, E. Oberla, B. Oeyen, N. Punsuebsay, L. Pyras, M. Ravn, A. Rifaie, D. Ryckbosch, O. Schlemper, F. Schl\"uter, O. Scholten, D. Seckel, M. F. H. Seikh, J. Stachurska, J. Stoffels, S. Toscano, D. Tosi, J. Tutt, D. J. Van Den Broeck, N. van Eijndhoven, A. G. Vieregg, A. Vijai, C. Welling, D. R. Williams, P. Windischhofer, S. Wissel, R. Young, A. Zink | Radio emission from airplanes as observed with RNO-G | null | null | null | null | hep-ex astro-ph.HE astro-ph.IM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper describes how intentional and unintentional radio emission from airplanes is recorded with the Radio Neutrino Observatory Greenland (RNO-G). We characterize the received signals and define a procedure to extract a clean set of impulsive signals. These signals are highly suitable for instrument calibration, also for future experiments. A set of signals is used to probe the timing precision of RNO-G in-situ, which is found to match expectations. We also discuss the impact of these signals on the ability to detect neutrinos with RNO-G. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 00:09:49 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"G Collaboration",
"",
""
],
[
"Agarwal",
"S.",
""
],
[
"Aguilar",
"J. A.",
""
],
[
"Alden",
"N.",
""
],
[
"Ali",
"S.",
""
],
[
"Allison",
"P.",
""
],
[
"Betts",
"M.",
""
],
[
"Besson",
"D.",
""
],
[
"Bishop",
"A.",
""
],
[
"Botner",
"O.",
""
],
[
"Bouma",
"S.",
""
],
[
"Buitink",
"S.",
""
],
[
"Camphyn",
"R.",
""
],
[
"Chan",
"J.",
""
],
[
"Chiche",
"S.",
""
],
[
"Clark",
"B. A.",
""
],
[
"Coleman",
"A.",
""
],
[
"Couberly",
"K.",
""
],
[
"de Kockere",
"S.",
""
],
[
"de Vries",
"K. D.",
""
],
[
"Deaconu",
"C.",
""
],
[
"Giri",
"P.",
""
],
[
"Glaser",
"C.",
""
],
[
"Glüsenkamp",
"T.",
""
],
[
"Gui",
"H.",
""
],
[
"Hallgren",
"A.",
""
],
[
"Hallmann",
"S.",
""
],
[
"Hanson",
"J. C.",
""
],
[
"Helbing",
"K.",
""
],
[
"Hendricks",
"B.",
""
],
[
"Henrichs",
"J.",
""
],
[
"Heyer",
"N.",
""
],
[
"Hornhuber",
"C.",
""
],
[
"Santiago",
"E. Huesca",
""
],
[
"Hughes",
"K.",
""
],
[
"Jaitly",
"A.",
""
],
[
"Karg",
"T.",
""
],
[
"Karle",
"A.",
""
],
[
"Kelley",
"J. L.",
""
],
[
"Kimo",
"J.",
""
],
[
"Kopper",
"C.",
""
],
[
"Korntheuer",
"M.",
""
],
[
"Kowalski",
"M.",
""
],
[
"Kravchenko",
"I.",
""
],
[
"Krebs",
"R.",
""
],
[
"Kugelmeier",
"M.",
""
],
[
"Lahmann",
"R.",
""
],
[
"Liu",
"C. -H.",
""
],
[
"Marsee",
"M. J.",
""
],
[
"Meyers",
"Z. S.",
""
],
[
"Mulrey",
"K.",
""
],
[
"Muzio",
"M.",
""
],
[
"Nelles",
"A.",
""
],
[
"Novikov",
"A.",
""
],
[
"Nozdrina",
"A.",
""
],
[
"Oberla",
"E.",
""
],
[
"Oeyen",
"B.",
""
],
[
"Punsuebsay",
"N.",
""
],
[
"Pyras",
"L.",
""
],
[
"Ravn",
"M.",
""
],
[
"Rifaie",
"A.",
""
],
[
"Ryckbosch",
"D.",
""
],
[
"Schlemper",
"O.",
""
],
[
"Schlüter",
"F.",
""
],
[
"Scholten",
"O.",
""
],
[
"Seckel",
"D.",
""
],
[
"Seikh",
"M. F. H.",
""
],
[
"Stachurska",
"J.",
""
],
[
"Stoffels",
"J.",
""
],
[
"Toscano",
"S.",
""
],
[
"Tosi",
"D.",
""
],
[
"Tutt",
"J.",
""
],
[
"Broeck",
"D. J. Van Den",
""
],
[
"van Eijndhoven",
"N.",
""
],
[
"Vieregg",
"A. G.",
""
],
[
"Vijai",
"A.",
""
],
[
"Welling",
"C.",
""
],
[
"Williams",
"D. R.",
""
],
[
"Windischhofer",
"P.",
""
],
[
"Wissel",
"S.",
""
],
[
"Young",
"R.",
""
],
[
"Zink",
"A.",
""
]
] |
2506.17532 | Bennett Sessa | Bennett C. Sessa, Federico Cao, Robert A. Pelcovits, Thomas R. Powers, Guillaume Duclos | Interfacial instability of confined 3D active droplets | null | null | null | null | cond-mat.soft | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Instabilities of fluid-fluid interfaces are ubiquitous in passive soft matter. Adding activity to the interface or either fluid can dramatically change the stability of the interface. Using experiment and theory, we investigate the interfacial instability of a deformable 3D active nematic liquid crystal droplet in the isotropic phase surrounded by a passive fluid and confined between two parallel plates. Spontaneous active flows drive the growth of undulations along the active/passive interface, with the mode number of the fastest-growing mode increasing with droplet radius and decreasing with gap height. We apply the lubrication approximation to a minimal nematohydrodynamic model to determine the growth rates of all interfacial modes. The magnitude of the growth rate is determined by the active timescale and the relaxation timescales associated with liquid crystalline order, as well as capillary and viscous stresses. We find multiple points of agreement between experiment and theory, including the shape evolution of individual droplets, the growth rates of unstable modes averaged across many droplets, and the extensional shear flows observed within droplets. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 01:06:28 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 17:40:52 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Sessa",
"Bennett C.",
""
],
[
"Cao",
"Federico",
""
],
[
"Pelcovits",
"Robert A.",
""
],
[
"Powers",
"Thomas R.",
""
],
[
"Duclos",
"Guillaume",
""
]
] |
2506.17539 | Sidong Feng | Sidong Feng, Changhao Du, Huaxiao Liu, Qingnan Wang, Zhengwei Lv, Mengfei Wang, Chunyang Chen | Breaking Single-Tester Limits: Multi-Agent LLMs for Multi-User Feature Testing | Accepted to International Conference on Software Engineering (ICSE 2026). arXiv admin note: substantial text overlap with arXiv:2504.15474 | null | null | null | cs.SE | http://creativecommons.org/licenses/by/4.0/ | The growing dependence on mobile phones and their apps has made multi-user interactive features, like chat calls, live streaming, and video conferencing, indispensable for bridging the gaps in social connectivity caused by physical and situational barriers. However, automating these interactive features for testing is fraught with challenges, owing to their inherent need for timely, dynamic, and collaborative user interactions, which current automated testing methods inadequately address. Inspired by the concept of agents designed to autonomously and collaboratively tackle problems, we propose MAdroid, a novel multi-agent approach powered by the Large Language Models (LLMs) to automate the multi-user interactive task for app feature testing. Specifically, MAdroid employs two functional types of multi-agents: user agents (Operator) and supervisor agents (Coordinator and Observer). Each agent takes a specific role: the Coordinator directs the interactive task; the Operator mimics user interactions on the device; and the Observer monitors and reviews the task automation process. Our evaluation, which included 41 multi-user interactive tasks, demonstrates the effectiveness of our approach, achieving 82.9% of the tasks with 96.8% action similarity, outperforming the ablation studies and state-of-the-art baselines. Additionally, a preliminary investigation underscores MAdroid's practicality by helping identify 11 multi-user interactive bugs during regression app testing, confirming its potential value in real-world software development contexts. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 01:38:53 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 00:54:08 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Feng",
"Sidong",
""
],
[
"Du",
"Changhao",
""
],
[
"Liu",
"Huaxiao",
""
],
[
"Wang",
"Qingnan",
""
],
[
"Lv",
"Zhengwei",
""
],
[
"Wang",
"Mengfei",
""
],
[
"Chen",
"Chunyang",
""
]
] |
2506.17542 | Nitin Venkateswaran | Nitin Venkateswaran and Kevin Tang and Ratree Wayland | Probing for Phonology in Self-Supervised Speech Representations: A Case Study on Accent Perception | null | null | null | null | cs.SD cs.CL eess.AS | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Traditional models of accent perception underestimate the role of gradient variations in phonological features which listeners rely upon for their accent judgments. We investigate how pretrained representations from current self-supervised learning (SSL) models of speech encode phonological feature-level variations that influence the perception of segmental accent. We focus on three segments: the labiodental approximant, the rhotic tap, and the retroflex stop, which are uniformly produced in the English of native speakers of Hindi as well as other languages in the Indian sub-continent. We use the CSLU Foreign Accented English corpus (Lander, 2007) to extract, for these segments, phonological feature probabilities using Phonet (V\'asquez-Correa et al., 2019) and pretrained representations from Wav2Vec2-BERT (Barrault et al., 2023) and WavLM (Chen et al., 2022) along with accent judgements by native speakers of American English. Probing analyses show that accent strength is best predicted by a subset of the segment's pretrained representation features, in which perceptually salient phonological features that contrast the expected American English and realized non-native English segments are given prominent weighting. A multinomial logistic regression of pretrained representation-based segment distances from American and Indian English baselines on accent ratings reveals strong associations between the odds of accent strength and distances from the baselines, in the expected directions. These results highlight the value of self-supervised speech representations for modeling accent perception using interpretable phonological features. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 01:44:19 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Venkateswaran",
"Nitin",
""
],
[
"Tang",
"Kevin",
""
],
[
"Wayland",
"Ratree",
""
]
] |
2506.17548 | Utkarsh Chadha | Lydia Lonardi, Caitlyn Lew-Tong, Brian Dunsten Miranda, Chandraleka V.S., Evelyn (Wenxuan) Li, Bhaavyaa Sahukari, Harshit Poddar, Keerthana Satheesh, and Utkarsh Chadha | Toward a Circular Nanotechnology for Biofuels: Integrating Sustainable Synthesis, Recovery, and Performance Optimization | Currently Submitted to "Advanced Sustainable Systems" | null | null | null | cond-mat.mtrl-sci physics.app-ph | http://creativecommons.org/licenses/by/4.0/ | This review exhaustively evaluates the role of nanomaterials across synthesis, characterization and application stages of biofuel systems. The common types of nanomaterials that are used for biofuel applications include metal oxides, carbon-based structures, and hybrids are evaluated for their effectiveness in efficient biofuel production. The properties of such nanomaterials are being utilized as an aid to produce biofuels through improved catalysis, enzyme immobilization and thermal stability. Common synthesis methods such as sol-gel, co-precipitation, and green synthesis are compared, alongside characterization tools like TEM, SEM, FTIR, and BET. This study focuses on transesterification, biomass pretreatment, and fermentation processes, where nanomaterials significantly improve yield and reusability. There are several challenges despite the merit of using nanomaterials, the tradeoffs include costs, scalability, and environmental impact, which further expands into evaluating the life cycle of such materials being used. This review outlines the practical potential of the nanomaterials in enabling more efficient and sustainable biofuel production. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 02:35:25 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Lonardi",
"Lydia",
"",
"Wenxuan"
],
[
"Lew-Tong",
"Caitlyn",
"",
"Wenxuan"
],
[
"Miranda",
"Brian Dunsten",
"",
"Wenxuan"
],
[
"S.",
"Chandraleka V.",
"",
"Wenxuan"
],
[
"Evelyn",
"",
"",
"Wenxuan"
],
[
"Li",
"",
""
],
[
"Sahukari",
"Bhaavyaa",
""
],
[
"Poddar",
"Harshit",
""
],
[
"Satheesh",
"Keerthana",
""
],
[
"Chadha",
"Utkarsh",
""
]
] |
2506.17551 | Haowei Yang | Haowei Yang, Yu Tian, Zhongheng Yang, Zhao Wang, Chengrui Zhou, Dannier Li | Research on Model Parallelism and Data Parallelism Optimization Methods in Large Language Model-Based Recommendation Systems | null | null | null | null | cs.DC cs.AI | http://creativecommons.org/licenses/by/4.0/ | With the rapid adoption of large language models (LLMs) in recommendation systems, the computational and communication bottlenecks caused by their massive parameter sizes and large data volumes have become increasingly prominent. This paper systematically investigates two classes of optimization methods-model parallelism and data parallelism-for distributed training of LLMs in recommendation scenarios. For model parallelism, we implement both tensor parallelism and pipeline parallelism, and introduce an adaptive load-balancing mechanism to reduce cross-device communication overhead. For data parallelism, we compare synchronous and asynchronous modes, combining gradient compression and sparsification techniques with an efficient aggregation communication framework to significantly improve bandwidth utilization. Experiments conducted on a real-world recommendation dataset in a simulated service environment demonstrate that our proposed hybrid parallelism scheme increases training throughput by over 30% and improves resource utilization by approximately 20% compared to traditional single-mode parallelism, while maintaining strong scalability and robustness. Finally, we discuss trade-offs among different parallel strategies in online deployment and outline future directions involving heterogeneous hardware integration and automated scheduling technologies. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 02:37:25 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 02:28:50 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Yang",
"Haowei",
""
],
[
"Tian",
"Yu",
""
],
[
"Yang",
"Zhongheng",
""
],
[
"Wang",
"Zhao",
""
],
[
"Zhou",
"Chengrui",
""
],
[
"Li",
"Dannier",
""
]
] |
2506.17559 | Jingjing Cui | Enzhi Zhou, Jingjing Cui, and Ziyue Liu, Zhiguo Ding, and Pingzhi Fan | Joint Transmission for Cellular Networks with Pinching Antennas: System Design and Analysis | null | null | null | null | cs.IT eess.SP math.IT | http://creativecommons.org/licenses/by-nc-nd/4.0/ | As an emerging flexible antenna technology for wireless communications, pinching-antenna systems, offer distinct advantages in terms of cost efficiency and deployment flexibility. This paper investigates joint transmission strategies of the base station (BS) and pinching antennas (PAS), focusing specifically on how to cooperate efficiently between the BS and waveguide-mounted pinching antennas for enhancing the performance of the user equipment (UE). By jointly considering the performance, flexibility, and complexity, we propose three joint BS-PAS transmission schemes along with the best beamforming designs, namely standalone deployment (SD), semi-cooperative deployment (SCD) and full-cooperative deployment (FCD). More specifically, for each BS-PAS joint transmission scheme, we conduct a comprehensive performance analysis in terms of the power allocation strategy, beamforming design, and practical implementation considerations. We also derive closed-form expressions for the average received SNR across the proposed BS-PAS joint transmission schemes, which are verified through Monte Carlo simulations. Finally, numerical results demonstrate that deploying pinching antennas in cellular networks, particularly through cooperation between the BS and PAS, can achieve significant performance gains. We further identify and characterize the key network parameters that influence the performance, providing insights for deploying pinching antennas. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 03:03:04 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Zhou",
"Enzhi",
""
],
[
"Cui",
"Jingjing",
""
],
[
"Liu",
"Ziyue",
""
],
[
"Ding",
"Zhiguo",
""
],
[
"Fan",
"Pingzhi",
""
]
] |
2506.17570 | Wei Sun | Wei Sun, Minghong Fang, Mengyuan Li | VReaves: Eavesdropping on Virtual Reality App Identity and Activity via Electromagnetic Side Channels | null | null | null | null | cs.NI cs.CR cs.HC | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Virtual reality (VR) has recently proliferated significantly, consisting of headsets or head-mounted displays (HMDs) and hand controllers for an embodied and immersive experience. The VR device is usually embedded with different kinds of IoT sensors, such as cameras, microphones, communication sensors, etc. However, VR security has not been scrutinized from a physical hardware point of view, especially electromagnetic emanations (EM) that are automatically and unintentionally emitted from the VR headset. This paper presents VReaves, a system that can eavesdrop on the electromagnetic emanation side channel of a VR headset for VR app identification and activity recognition. To do so, we first characterize the electromagnetic emanations from the embedded IoT sensors (e.g., cameras and microphones) in the VR headset through a signal processing pipeline and further propose machine learning models to identify the VR app and recognize the VR app activities. Our experimental evaluation with commercial off-the-shelf VR devices demonstrates the efficiency of VR app identification and activity recognition via electromagnetic emanation side channel. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 03:53:24 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 03:33:43 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Sun",
"Wei",
""
],
[
"Fang",
"Minghong",
""
],
[
"Li",
"Mengyuan",
""
]
] |
2506.17611 | Jinchuan Tian | Jinchuan Tian, William Chen, Yifan Peng, Jiatong Shi, Siddhant Arora, Shikhar Bharadwaj, Takashi Maekaku, Yusuke Shinohara, Keita Goto, Xiang Yue, Huck Yang, Shinji Watanabe | OpusLM: A Family of Open Unified Speech Language Models | null | null | null | null | cs.CL cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | This paper presents Open Unified Speech Language Models (OpusLMs), a family of open foundational speech language models (SpeechLMs) up to 7B. Initialized from decoder-only text language models, the OpusLMs are continuously pre-trained on 213K hours of speech-text pairs and 292B text-only tokens. We demonstrate our OpusLMs achieve comparable (or even superior) performance with existing SpeechLMs in speech recognition, speech synthesis, and text-only capabilities. Technically, this paper articulates our SpeechLM designs on tokenization, multi-stream language models, and multi-stage training strategies. We experimentally demonstrate the importance of model size scaling and the effect of annealing data selection. The OpusLMs are all built from publicly available materials and are fully transparent models. We release our code, data, checkpoints, and training logs to facilitate open SpeechLM research | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 06:30:59 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Tian",
"Jinchuan",
""
],
[
"Chen",
"William",
""
],
[
"Peng",
"Yifan",
""
],
[
"Shi",
"Jiatong",
""
],
[
"Arora",
"Siddhant",
""
],
[
"Bharadwaj",
"Shikhar",
""
],
[
"Maekaku",
"Takashi",
""
],
[
"Shinohara",
"Yusuke",
""
],
[
"Goto",
"Keita",
""
],
[
"Yue",
"Xiang",
""
],
[
"Yang",
"Huck",
""
],
[
"Watanabe",
"Shinji",
""
]
] |
2506.17651 | Jiaxu Li | Jiaxu Li, Zijian Zhang, Shiqi Wang, Yu He, Haochang Lyu, Qiusha Wang, Bowen Dong, Daoqian Zhu, Hisakazu Matsuki, Dapeng Zhu, Guang Yang, Weisheng Zhao | Field-Free Superconducting Diode Enabled by Geometric Asymmetry and Perpendicular Magnetization | null | null | null | null | cond-mat.supr-con | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The superconducting diode effect (SDE)- manifested as directional, dissipationless supercurrents - is pivotal for realizing energy-efficient superconducting logic and memory technologies. Achieving high-efficiency SDE without external magnetic fields, however, remains a fundamental challenge. Here, we report a strongly enhanced, field-free SDE in Pt/Co/Nb heterostructures, enabled by the interplay of engineered geometric asymmetry and stray fields from a perpendicularly magnetized Co layer. This configuration promotes directional vortex entry and spatially selective pinning, yielding diode efficiencies that exceed all previously reported field-free values. Temperature- and field-dependent transport measurements, supported by micromagnetic simulations, reveal that the enhanced nonreciprocity stems from three cooperative mechanisms: asymmetric vortex entry, localized magnetic pinning, and Lorentz-force imbalance. These findings establish a scalable, CMOS-compatible platform for high-performance superconducting rectifiers, offering new opportunities for cryogenic spintronics and quantum electronics. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 09:21:44 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 04:33:05 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Li",
"Jiaxu",
""
],
[
"Zhang",
"Zijian",
""
],
[
"Wang",
"Shiqi",
""
],
[
"He",
"Yu",
""
],
[
"Lyu",
"Haochang",
""
],
[
"Wang",
"Qiusha",
""
],
[
"Dong",
"Bowen",
""
],
[
"Zhu",
"Daoqian",
""
],
[
"Matsuki",
"Hisakazu",
""
],
[
"Zhu",
"Dapeng",
""
],
[
"Yang",
"Guang",
""
],
[
"Zhao",
"Weisheng",
""
]
] |
2506.17655 | Senol Gulgonul | Senol Gulgonul | PID Tuning via Desired Step Response Curve Fitting | 4 tables, 4 figures | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | This paper presents a PID tuning method based on step response curve fitting (PID-SRCF) that utilizes L2-norm minimization for precise reference tracking and explicit transient response shaping. The algorithm optimizes controller parameters by minimizing the root-mean-square error between desired and actual step responses. The proposed approach determines optimal PID parameters by matching any closed-loop response to a desired system step response. Practically a first-order plus time delay model or a second-order system with defined settling time and overshoot requirements are preferred. The method has open-source implementation using constrained nonlinear optimization in MATLAB. Comparative evaluations demonstrate that PID-SRCF can replace known analytical methods like Ziegler Nichols, Lambda Tuning, Pole Placement, Dominant Pole and MATLAB proprietary PID tuning applications. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 09:25:43 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 17:34:03 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Gulgonul",
"Senol",
""
]
] |
2506.17694 | Rajasekar Gnana Praveen | Gnana Praveen Rajasekhar, Jahangir Alam | SSAVSV: Towards Unified Model for Self-Supervised Audio-Visual Speaker Verification | null | null | null | null | cs.CV cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Conventional audio-visual methods for speaker verification rely on large amounts of labeled data and separate modality-specific architectures, which is computationally expensive, limiting their scalability. To address these problems, we propose a self-supervised learning framework based on contrastive learning with asymmetric masking and masked data modeling to obtain robust audiovisual feature representations. In particular, we employ a unified framework for self-supervised audiovisual speaker verification using a single shared backbone for audio and visual inputs, leveraging the versatility of vision transformers. The proposed unified framework can handle audio, visual, or audiovisual inputs using a single shared vision transformer backbone during training and testing while being computationally efficient and robust to missing modalities. Extensive experiments demonstrate that our method achieves competitive performance without labeled data while reducing computational costs compared to traditional approaches. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 12:02:53 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Rajasekhar",
"Gnana Praveen",
""
],
[
"Alam",
"Jahangir",
""
]
] |
2506.17703 | Ramazan Sahin | Elif Ozturk, Hira Asif, Mehmet Gunay, Mehmet Emre Tasgin, and Ramazan Sahin | Optically tunable linear and nonlinear enhancement of index of refraction | 6 pages, 5 figures | null | null | null | physics.optics physics.app-ph quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Control of optical properties of materials by tuning their refractive index can revolutionize the current state-of-the-art technology to manipulate light propagation in the high loss media. Here we demonstrate active optical tuning of the plasmonic analog of \textit{enhancement of index of refraction} (EIR) in both linear and nonlinear regimes using a quantum mechanical approach. By employing a pump-probe scheme, we investigate the tuning of refractive index of the probe field by varying amplitude and phase of the pump source. In contrast to classical approach used in \cite{Panahpour2019}, we formulate both first- and second-order quantization to analyze nonlinear enhancement in the refractive index by modulating the response function of probe field. This approach enables indirect tuning of nonlinear modes and coherent control of the probe pulse under the coupling of linear plasmonic modes supported by two L-shaped nano-ellipsoids. Varying the pump amplitude not only shows a significant enhancement in the EIR in both regimes but also effectively suppresses optical losses with zero dispersion at the system's resonance frequency. Additionally, tuning pump phase induces a spectral shift in the frequency of the probe field which open new ways for active tuning of epsilon-near-zero (ENZ) materials. Our approach offers all-optical tuning of nonlinear refractive index which is essential for quantum technological applications. It also provides coherent control of optical properties of plasmonic nanostructures with applications in loss-compensated propagation and zero-index to high-refractive-index plasmonic metamaterials, as well as photonic switches. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 12:39:40 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 17:08:09 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Ozturk",
"Elif",
""
],
[
"Asif",
"Hira",
""
],
[
"Gunay",
"Mehmet",
""
],
[
"Tasgin",
"Mehmet Emre",
""
],
[
"Sahin",
"Ramazan",
""
]
] |
2506.17720 | Klaus Frahm | Klaus M. Frahm and Dima L. Shepelyansky | Wealth Thermalization Hypothesis | 14 pages (5 main and 9 SupMat), 6+12 figures | null | null | null | cond-mat.stat-mech q-fin.ST | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We introduce the wealth thermalization hypothesis according to which the wealth shared in a country or the whole world is described by the Rayleigh-Jeans thermal distribution with two conserved quantities of system wealth and norm or number of agents. This distribution depends on a dimensional parameter being the ratio of system total wealth and its dispersion range determined by highest revenues. At relatively small values of this ratio there is a formation of the Rayleigh-Jeans condensate, well studied in such physical systems as multimode optical fibers. This leads to a huge fraction of poor households and a small oligarchic fraction which monopolizes a dominant fraction of total wealth thus generating a strong inequality in human society. We show that this thermalization gives a good description of real data of Lorenz curves of US, UK, the whole world and capitalization of S\&P500 companies at New York Stock Exchange. Possible actions for inequality reduction are briefly discussed. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 14:19:59 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Frahm",
"Klaus M.",
""
],
[
"Shepelyansky",
"Dima L.",
""
]
] |
2506.17725 | Peter Piechulla | Peter M. Piechulla, Mingliang Chen, Aristeidis Goulas, Riikka L. Puurunen, J. Ruud van Ommen | Atomic layer deposition on particulate materials from 1988 through 2023: A quantitative review of technologies, materials and applications | Supporting Information at the end of the manuscript; to be published in ACS Chemistry of Materials | null | null | null | cond-mat.mtrl-sci physics.app-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Atomic layer deposition (ALD) is widely studied for numerous applications and is commercially employed in the semiconductor industry, where planar substrates are the norm. However, the inherent ALD feature of coating virtually any surface geometry with atomistic thickness control is equally attractive for coating particulate materials (supports). In this review, we provide a comprehensive overview of the developments in this decades-old field of ALD on particulate materials, drawing on a bottom-up and quantitative analysis of 799 articles from this field. The obtained dataset is the basis for abstractions regarding reactor types (specifically for particles), coating materials, reactants, supports and processing conditions. Furthermore, the dataset enables direct access to specific processing conditions (for a given material, surface functionality, application etc.) and increases accessibility of the respective literature. We also review fundamental concepts of ALD on particles, and discuss the most common applications, i.e., catalysis (thermo-, electro-, photo-), batteries, luminescent phosphors and healthcare. Finally, we identify historical trends, and provide an outlook on prospective developments. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 14:48:04 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Piechulla",
"Peter M.",
""
],
[
"Chen",
"Mingliang",
""
],
[
"Goulas",
"Aristeidis",
""
],
[
"Puurunen",
"Riikka L.",
""
],
[
"van Ommen",
"J. Ruud",
""
]
] |
2506.17728 | Jun Xu | Dalong Zhang, Jun Xu, Jun Zhou, Lei Liang, Lin Yuan, Ling Zhong, Mengshu Sun, Peilong Zhao, QiWei Wang, Xiaorui Wang, Xinkai Du, YangYang Hou, Yu Ao, ZhaoYang Wang, Zhengke Gui, ZhiYing Yi, Zhongpu Bo | KAG-Thinker: Interactive Thinking and Deep Reasoning in LLMs via Knowledge-Augmented Generation | null | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we introduce KAG-Thinker, which upgrade KAG to a multi-turn interactive thinking and deep reasoning framework powered by a dedicated parameter-light large language model (LLM). Our approach constructs a structured thinking process for solving complex problems, enhancing the the logical coherence and contextual consistency of the reasoning process in question-answering (Q&A) tasks on domain-specific knowledge bases (KBs) within LLMs. Following the \textbf{Logical Form} guided retrieval and reasoning technology route of KAG, this framework first decomposes complex questions into independently solvable sub-problems (which are also referred to as logical forms) through \textbf{breadth decomposition}. Each such logical form is represented in two equivalent forms-natural language and logical function-and subsequently classified as either a Knowledge Retrieval or Reasoning Analysis task. Dependencies and parameter passing between these tasks are explicitly modeled via logical function interfaces. In the solving process, the Retrieval function performs retrieval tasks. It retrieves one-hop structured and unstructured information of specified knowledge unit. While the Math and Deduce functions are used to perform reasoning analysis tasks. Secondly, it is worth noting that, in the Knowledge Retrieval sub-problem tasks, LLMs and external knowledge sources are regarded as equivalent KBs. We use the \textbf{knowledge boundary} module to determine the optimal source using self-regulatory mechanisms such as confidence calibration and reflective reasoning, and use the \textbf{depth solving} module to enhance the comprehensiveness of knowledge acquisition... | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 14:58:53 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 12:50:57 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Zhang",
"Dalong",
""
],
[
"Xu",
"Jun",
""
],
[
"Zhou",
"Jun",
""
],
[
"Liang",
"Lei",
""
],
[
"Yuan",
"Lin",
""
],
[
"Zhong",
"Ling",
""
],
[
"Sun",
"Mengshu",
""
],
[
"Zhao",
"Peilong",
""
],
[
"Wang",
"QiWei",
""
],
[
"Wang",
"Xiaorui",
""
],
[
"Du",
"Xinkai",
""
],
[
"Hou",
"YangYang",
""
],
[
"Ao",
"Yu",
""
],
[
"Wang",
"ZhaoYang",
""
],
[
"Gui",
"Zhengke",
""
],
[
"Yi",
"ZhiYing",
""
],
[
"Bo",
"Zhongpu",
""
]
] |
2506.17730 | Faxian Xiu | Chang Jiang, Fan Yang, Jinshan Yang, Peng Yu, Huiying Liu, Yuda Zhang, Zehao Jia, Xiangyu Cao, Jingyi Yan, Zheng Liu, Xian-Lei Sheng, Cong Xiao, Shengyuan A. Yang, Shaoming Dong, and Faxian Xiu | Room-temperature intrinsic nonlinear planar Hall effect in TaIrTe$_4$ | null | null | null | null | cond-mat.mes-hall cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Intrinsic responses are of paramount importance in physics research, as they represent the inherent properties of materials, independent of extrinsic factors that vary from sample to sample, and often reveal the intriguing quantum geometry of the band structure. Here, we report the experimental discovery of a new intrinsic response in charge transport, specifically the intrinsic nonlinear planar Hall effect (NPHE), in the topological semimetal TaIrTe$_4$. This effect is characterized by an induced Hall current that is quadratic in the driving electric field and linear in the in-plane magnetic field. The response coefficient is determined by the susceptibility tensor of Berry-connection polarizability dipole, which is an intrinsic band geometric quantity. Remarkably, the signal persists up to room temperature. Our theoretical calculations show excellent agreement with the experimental results and further elucidate the significance of a previously unknown orbital mechanism in intrinsic NPHE. This finding not only establishes a novel intrinsic material property but also opens a new route toward innovative nonlinear devices capable of operating at room temperature. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 15:09:14 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Jiang",
"Chang",
""
],
[
"Yang",
"Fan",
""
],
[
"Yang",
"Jinshan",
""
],
[
"Yu",
"Peng",
""
],
[
"Liu",
"Huiying",
""
],
[
"Zhang",
"Yuda",
""
],
[
"Jia",
"Zehao",
""
],
[
"Cao",
"Xiangyu",
""
],
[
"Yan",
"Jingyi",
""
],
[
"Liu",
"Zheng",
""
],
[
"Sheng",
"Xian-Lei",
""
],
[
"Xiao",
"Cong",
""
],
[
"Yang",
"Shengyuan A.",
""
],
[
"Dong",
"Shaoming",
""
],
[
"Xiu",
"Faxian",
""
]
] |
2506.17742 | Phani Kumar Peddibhotla Dr. | Mayana Yousuf Ali Khan, Pralekh Dubey, Lakshmi Madhuri P, Ashutosh Kumar Tripathi, Phani Kumar Peddibhotla, and Pydi Ganga Bahubalindruni | Sensing Electric Currents in an a-IGZO TFT-Based Circuit Using a Quantum Diamond Microscope | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Quantum Diamond Microscope (QDM) is an emerging magnetic imaging tool enabling noninvasive characterization of electronic circuits through spatially mapping current densities. In this work, we demonstrate wafer-level current sensing of a current mirror circuit composed of 16 amorphous-indium-gallium-zinc oxide (a-IGZO) thin-film transistors (TFTs). a-IGZO TFTs are promising for flexible electronics due to their high performance. Using QDM, we obtain two-dimensional (2D) magnetic field images produced by DC currents, from which accurate current density maps are extracted. Notably, QDM measurements agree well with conventional electrical probing measurements, and enable current sensing in internal circuit paths inaccessible via conventional methods. Our results highlight QDM's capability as a noninvasive diagnostic tool for the characterization of emerging semiconductor technologies, especially oxide-based TFTs. This approach provides essential insights to fabrication engineers, with potential to improve yield and reliability in flexible electronics manufacturing. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 15:38:58 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 05:56:02 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Khan",
"Mayana Yousuf Ali",
""
],
[
"Dubey",
"Pralekh",
""
],
[
"P",
"Lakshmi Madhuri",
""
],
[
"Tripathi",
"Ashutosh Kumar",
""
],
[
"Peddibhotla",
"Phani Kumar",
""
],
[
"Bahubalindruni",
"Pydi Ganga",
""
]
] |
2506.17789 | N J Karthika | N J Karthika, Maharaj Brahma, Rohit Saluja, Ganesh Ramakrishnan, Maunendra Sankar Desarkar | Multilingual Tokenization through the Lens of Indian Languages: Challenges and Insights | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Tokenization plays a pivotal role in multilingual NLP. However, existing tokenizers are often skewed towards high-resource languages, limiting their effectiveness for linguistically diverse and morphologically rich languages such as those in the Indian subcontinent. This paper presents a comprehensive intrinsic evaluation of tokenization strategies across 17 Indian languages. We quantify the trade-offs between bottom-up and top-down tokenizer algorithms (BPE and Unigram LM), effects of vocabulary sizes, and compare strategies of multilingual vocabulary construction such as joint and cluster-based training. We also show that extremely low-resource languages can benefit from tokenizers trained on related high-resource languages. Our study provides practical insights for building more fair, efficient, and linguistically informed tokenizers for multilingual NLP. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 18:47:33 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 09:35:36 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Karthika",
"N J",
""
],
[
"Brahma",
"Maharaj",
""
],
[
"Saluja",
"Rohit",
""
],
[
"Ramakrishnan",
"Ganesh",
""
],
[
"Desarkar",
"Maunendra Sankar",
""
]
] |
2506.17797 | Aryan Iliat | Aryan Iliat, Mark Byrd, Sahel Ashhab, and LianAo Wu | Physically motivated decompositions of single qutrit gates | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Although only two quantum states of a physical system are often used to encode quantum information in the form of qubits, many levels can in principle be used to obtain qudits and increase the information capacity of the system. To take advantage of the additional levels, a parameterization of unitary transformations in terms of experimentally realizable operations is needed. Many parameterizations of unitary 3 * 3 matrices (U(3)) exist. One decomposition of a general unitary matrix can be expressed as the product of an exponential of a diagonal matrix and an exponential of an off-diagonal matrix. This decomposition is relevant for controlling superconducting qutrits using fixed-frequency resonant control pulses. This decomposition is numerically confirmed to allow the parameterization of any element in U(3). It is shown that a simple setting of parameter ranges of parameters can easily lead to an over-parameterization, in the sense that several different sets of values for the parameters produce the same element in U(3). This fact is demonstrated using the Walsh-Hadamard (WH) matrix as an example, which is also a special qutrit gate of practical interest. The different decompositions are shown to be related, and the relationships between them are presented using general methods. The shortest path needed for the implementation of a qutrit gate is found. Other parameterizations obtained by other analytic means, which can be advantageous for various reasons, are also discussed. | [
{
"version": "v1",
"created": "Sat, 21 Jun 2025 19:40:21 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Iliat",
"Aryan",
""
],
[
"Byrd",
"Mark",
""
],
[
"Ashhab",
"Sahel",
""
],
[
"Wu",
"LianAo",
""
]
] |
2506.17879 | Zheng Chen | Zheng Chen | StainPIDR: A Pathological Image Decouplingand Reconstruction Method for Stain Normalization Based on Color Vector Quantization and Structure Restaining | null | null | null | null | eess.IV cs.AI cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The color appearance of a pathological image is highly related to the imaging protocols, the proportion of different dyes, and the scanning devices. Computer-aided diagnostic systems may deteriorate when facing these color-variant pathological images. In this work, we propose a stain normalization method called StainPIDR. We try to eliminate this color discrepancy by decoupling the image into structure features and vector-quantized color features, restaining the structure features with the target color features, and decoding the stained structure features to normalized pathological images. We assume that color features decoupled by different images with the same color should be exactly the same. Under this assumption, we train a fixed color vector codebook to which the decoupled color features will map. In the restaining part, we utilize the cross-attention mechanism to efficiently stain the structure features. As the target color (decoupled from a selected template image) will also affect the performance of stain normalization, we further design a template image selection algorithm to select a template from a given dataset. In our extensive experiments, we validate the effectiveness of StainPIDR and the template image selection algorithm. All the results show that our method can perform well in the stain normalization task. The code of StainPIDR will be publicly available later. | [
{
"version": "v1",
"created": "Sun, 22 Jun 2025 02:54:20 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Chen",
"Zheng",
""
]
] |
2506.17885 | Trong-An Bui Dr. | Trong-An Bui, Thanh-Thoai Le | Cloud-Aware SAR Fusion for Enhanced Optical Sensing in Space Missions | null | null | null | null | cs.CV cs.LG eess.IV | http://creativecommons.org/licenses/by/4.0/ | Cloud contamination significantly impairs the usability of optical satellite imagery, affecting critical applications such as environmental monitoring, disaster response, and land-use analysis. This research presents a Cloud-Attentive Reconstruction Framework that integrates SAR-optical feature fusion with deep learning-based image reconstruction to generate cloud-free optical imagery. The proposed framework employs an attention-driven feature fusion mechanism to align complementary structural information from Synthetic Aperture Radar (SAR) with spectral characteristics from optical data. Furthermore, a cloud-aware model update strategy introduces adaptive loss weighting to prioritize cloud-occluded regions, enhancing reconstruction accuracy. Experimental results demonstrate that the proposed method outperforms existing approaches, achieving a PSNR of 31.01 dB, SSIM of 0.918, and MAE of 0.017. These outcomes highlight the framework's effectiveness in producing high-fidelity, spatially and spectrally consistent cloud-free optical images. | [
{
"version": "v1",
"created": "Sun, 22 Jun 2025 03:27:41 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Bui",
"Trong-An",
""
],
[
"Le",
"Thanh-Thoai",
""
]
] |
2506.17886 | Julien Guinot | Julien Guinot, Elio Quinton, Gy\"orgy Fazekas | GD-Retriever: Controllable Generative Text-Music Retrieval with Diffusion Models | Accepted to ISMIR 2025 | null | null | null | cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | Multimodal contrastive models have achieved strong performance in text-audio retrieval and zero-shot settings, but improving joint embedding spaces remains an active research area. Less attention has been given to making these systems controllable and interactive for users. In text-music retrieval, the ambiguity of freeform language creates a many-to-many mapping, often resulting in inflexible or unsatisfying results.
We introduce Generative Diffusion Retriever (GDR), a novel framework that leverages diffusion models to generate queries in a retrieval-optimized latent space. This enables controllability through generative tools such as negative prompting and denoising diffusion implicit models (DDIM) inversion, opening a new direction in retrieval control. GDR improves retrieval performance over contrastive teacher models and supports retrieval in audio-only latent spaces using non-jointly trained encoders. Finally, we demonstrate that GDR enables effective post-hoc manipulation of retrieval behavior, enhancing interactive control for text-music retrieval tasks. | [
{
"version": "v1",
"created": "Sun, 22 Jun 2025 03:30:27 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 14:25:12 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Guinot",
"Julien",
""
],
[
"Quinton",
"Elio",
""
],
[
"Fazekas",
"György",
""
]
] |
2506.17898 | Dongdong Hu | Dongdong Hu | Cotorsion pairs in extensions of abelian categories | 21 pages | null | null | null | math.RT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $\mathcal{B}$ be an abelian category with enough projective objects and enough injective objects and let $\mathcal{A}=\mathcal{B}\ltimes_\eta\mathsf{F}$ be an $\eta$-extension of $\mathcal{B}$. Given a cotorsion pair $(\mathcal{X},\;\mathcal{Y})$ in $\mathcal{B}$, we construct a cotorsion pair $(^{\perp}{\mathsf{U}^{-1}(\mathcal{Y})}, \mathsf{U}^{-1}(\mathcal{Y}))$ in $\mathcal{A}$ and a cotorsion pair $(\Delta(\mathcal{X}),\;\Delta(\mathcal{X})^\perp)$ in $\mathcal{A}$ for $\mathsf{F}^2=0$. In addition, the heredity and completeness of these cotorsion pairs are studied. Finally, we give some applications and examples in comma categories, some Morita context rings and trivial extensions of rings. | [
{
"version": "v1",
"created": "Sun, 22 Jun 2025 04:34:46 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 01:52:44 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Hu",
"Dongdong",
""
]
] |
2506.17938 | Julius Fergy Rabago | Moustapha Essahraoui and Elmehdi Cherrat and Lekbir Afraites and Julius Fergy Tiongson Rabago | Simultaneous recovery of a corroded boundary and admittance using the Kohn-Vogelius method | in \textit{Computational Methods for Inverse Problems and Applications}, ICMDS 2024, Khouribga, Morocco, October 21--22, 2024, Springer Proc. Math. Stat., vol. 498, Springer, Cham, 2025 | null | null | null | math.NA cs.NA math.OC | http://creativecommons.org/licenses/by/4.0/ | We address the problem of identifying an unknown portion $\Gamma$ of the boundary of a $d$-dimensional ($d \in \{1, 2\}$) domain $\Omega$ and its associated Robin admittance coefficient, using two sets of boundary Cauchy data $(f, g)$--representing boundary temperature and heat flux--measured on the accessible portion $\Sigma$ of the boundary. Identifiability results \cite{Bacchelli2009,PaganiPierotti2009} indicate that a single measurement on $\Sigma$ is insufficient to uniquely determine both $\Gamma$ and $\alpha$, but two independent inputs yielding distinct solutions ensure the uniqueness of the pair $\Gamma$ and $\alpha$. In this paper, we propose a cost function based on the energy-gap of two auxiliary problems. We derive the variational derivatives of this objective functional with respect to both the Robin boundary $\Gamma$ and the admittance coefficient $\alpha$. These derivatives are utilized to develop a nonlinear gradient-based iterative scheme for the simultaneous numerical reconstruction of $\Gamma$ and $\alpha$. Numerical experiments are presented to demonstrate the effectiveness and practicality of the proposed method. | [
{
"version": "v1",
"created": "Sun, 22 Jun 2025 08:09:55 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 07:33:23 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Essahraoui",
"Moustapha",
""
],
[
"Cherrat",
"Elmehdi",
""
],
[
"Afraites",
"Lekbir",
""
],
[
"Rabago",
"Julius Fergy Tiongson",
""
]
] |
2506.17954 | Tan Chye Cheah | Liong Gele and Tan Chye Cheah | Mobile Image Analysis Application for Mantoux Skin Test | null | null | null | null | eess.IV cs.CV | http://creativecommons.org/licenses/by/4.0/ | This paper presents a newly developed mobile application designed to diagnose Latent Tuberculosis Infection (LTBI) using the Mantoux Skin Test (TST). Traditional TST methods often suffer from low follow-up return rates, patient discomfort, and subjective manual interpretation, particularly with the ball-point pen method, leading to misdiagnosis and delayed treatment. Moreover, previous developed mobile applications that used 3D reconstruction, this app utilizes scaling stickers as reference objects for induration measurement. This mobile application integrates advanced image processing technologies, including ARCore, and machine learning algorithms such as DeepLabv3 for robust image segmentation and precise measurement of skin indurations indicative of LTBI. The system employs an edge detection algorithm to enhance accuracy. The application was evaluated against standard clinical practices, demonstrating significant improvements in accuracy and reliability. This innovation is crucial for effective tuberculosis management, especially in resource-limited regions. By automating and standardizing TST evaluations, the application enhances the accessibility and efficiency of TB di-agnostics. Future work will focus on refining machine learning models, optimizing measurement algorithms, expanding functionalities to include comprehensive patient data management, and enhancing ARCore's performance across various lighting conditions and operational settings. | [
{
"version": "v1",
"created": "Sun, 22 Jun 2025 09:12:27 GMT"
}
] | 2025-06-25T00:00:00 | [
[
"Gele",
"Liong",
""
],
[
"Cheah",
"Tan Chye",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.