id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.14656 | Ziwei Hong | Ziwei Hong, Zhiyong Zheng | Twisted second moment of primitive cubic L-functions | null | null | null | null | math.NT | http://creativecommons.org/licenses/by/4.0/ | We investigate the mean value of the twisted second moment of primitive cubic $L$-functions over $\mathbb{F}_q(T)$ in the non-Kummer setting. Specifically, we study the sum
\begin{equation*}
\sum_{\substack{\chi\ primitive\ cubic\\ genus(\chi)=g}}\chi(h_1)\bar{\chi}(h_2)|L_q(\frac{1}{2}, \chi)|^2,
\end{equation*} where $L_q(s,\chi)$ denotes the $L$-function associated with primitive cubic character $\chi$. Employing a double Dirichlet series approach, we establish an error term of size | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:49:53 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Hong",
"Ziwei",
""
],
[
"Zheng",
"Zhiyong",
""
]
] |
2506.14657 | Woojoo Lee | Jongin Choi, Jina Park, Woojoo Lee, Jae-Jin Lee, Massoud Pedram | ASAP-FE: Energy-Efficient Feature Extraction Enabling Multi-Channel Keyword Spotting on Edge Processors | 7 pages, 11 figures, ISLPED 2025 | null | null | null | eess.AS cs.AR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multi-channel keyword spotting (KWS) has become crucial for voice-based applications in edge environments. However, its substantial computational and energy requirements pose significant challenges. We introduce ASAP-FE (Agile Sparsity-Aware Parallelized-Feature Extractor), a hardware-oriented front-end designed to address these challenges. Our framework incorporates three key innovations: (1) Half-overlapped Infinite Impulse Response (IIR) Framing: This reduces redundant data by approximately 25% while maintaining essential phoneme transition cues. (2) Sparsity-aware Data Reduction: We exploit frame-level sparsity to achieve an additional 50% data reduction by combining frame skipping with stride-based filtering. (3) Dynamic Parallel Processing: We introduce a parameterizable filter cluster and a priority-based scheduling algorithm that allows parallel execution of IIR filtering tasks, reducing latency and optimizing energy efficiency. ASAP-FE is implemented with various filter cluster sizes on edge processors, with functionality verified on FPGA prototypes and designs synthesized at 45 nm. Experimental results using TC-ResNet8, DS-CNN, and KWT-1 demonstrate that ASAP-FE reduces the average workload by 62.73% while supporting real-time processing for up to 32 channels. Compared to a conventional fully overlapped baseline, ASAP-FE achieves less than a 1% accuracy drop (e.g., 96.22% vs. 97.13% for DS-CNN), which is well within acceptable limits for edge AI. By adjusting the number of filter modules, our design optimizes the trade-off between performance and energy, with 15 parallel filters providing optimal performance for up to 25 channels. Overall, ASAP-FE offers a practical and efficient solution for multi-channel KWS on energy-constrained edge devices. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:50:24 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Choi",
"Jongin",
""
],
[
"Park",
"Jina",
""
],
[
"Lee",
"Woojoo",
""
],
[
"Lee",
"Jae-Jin",
""
],
[
"Pedram",
"Massoud",
""
]
] |
2506.14658 | Tianyu Yuan | Tianyu Yuan, Ivan Surovtsev, Megan C. King, Simon G. J. Mochrie | First-passage time to capture for diffusion in a 3D harmonic potential | 7 pages, 4 figures | null | null | null | math-ph math.MP physics.chem-ph physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | We determine the survival probability and first-passage time (FPT) to capture for a harmonically trapped particle, diffusing outside an absorbing spherical boundary by directly solving the differential equation for the survival probability. This solution, obtained as an infinite sum over the relevant eigenfunctions, corrects previously published results [D. S. Grebenkov, J. Phys. A 48, 013001 (2014)]. To verify our calculations, we perform simulations of the survival probability, that accurately reproduce the analytic solutions for a range of parameter values. We then obtain the corresponding FPT distribution as the negative time derivative of the survival probability. Finally, we derive an expression for mean first-passage time (MFPT), also as a sum over eigenfunctions. Numerical evaluation of the first twenty-five terms in this sum closely matches the MFPT obtained by a different method in D. S. Grebenkov, J. Phys. A 48, 013001 (2014). We also find that, in the limit of vanishing trap stiffness, the amplitude of the first term in our infinite-sum solution for the survival probability matches the theoretical escape probability for the potential-free diffusion-to-capture process. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:50:26 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Yuan",
"Tianyu",
""
],
[
"Surovtsev",
"Ivan",
""
],
[
"King",
"Megan C.",
""
],
[
"Mochrie",
"Simon G. J.",
""
]
] |
2506.14659 | Ariane Dekker | Ariane Dekker, Gonzalo Herrera and Dimitrios Kantzas | Axion-like particle limits from multi-messenger sources | 9 pages, 3 figures; Comments welcome | null | null | null | hep-ph astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | High-energy neutrino observation from the Seyfert galaxy NGC 1068 offers new insights into the non-thermal processes of active galactic nuclei. Simultaneous gamma-rays emitted by such sources can possibly oscillate into axion-like particles (ALPs) when propagating through astrophysical magnetic fields, potentially modifying the observed spectrum. To probe for ALP-induced signals, a robust understanding of the emission processes at the source is necessary. In this work, we perform a dedicated multi-messenger analysis by modeling a jet in the innermost vicinity of the central supermassive black hole of NGC 1068. We model in particular the neutrino and gamma-ray emission originating in lepto-hadronic collisions between jet accelerated particles and background particles from the corona, reproducing both the Fermi-LAT and IceCube data. These source models serve as a baseline for ALP searches, and we derive limits on the ALP-photon coupling by marginalizing over motivated ranges of astrophysical parameters. We find $g_{a\gamma} \lesssim 7 \times 10^{-11}$GeV$^{-1}$ for $m_a \lesssim 10^{-9}$ eV. These limits may be weaker than existing constraints, but they demonstrate the potential of multi-messenger observations to probe new physics. We conclude by discussing how additional upcoming multi-messenger sources and improved observational precision can enhance ALP sensitivity. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:50:32 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Dekker",
"Ariane",
""
],
[
"Herrera",
"Gonzalo",
""
],
[
"Kantzas",
"Dimitrios",
""
]
] |
2506.14660 | Alberto Mistroni | Alberto Mistroni, Marco Lisker, Yuji Yamamoto, Wei-Chen Wen, Fabian Fidorra, Henriette Tetzner, Laura K. Diebel, Lino Visser, Spandan Anupam, Vincent Mourik, Lars R. Schreiber, Hendrik Bluhm, Dominique Bougeard, Marvin H. Zoellner, Giovanni Capellini, Felix Reichmann | High yield, low disorder Si/SiGe heterostructures for spin qubit devices manufactured in a BiCMOS pilot line | 14 pages, 2 figures | null | null | null | cond-mat.mes-hall physics.app-ph | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The prospect of achieving fault-tolerant quantum computing with semiconductor spin qubits in Si/SiGe heterostructures relies on the integration of a large number of identical devices, a feat achievable through a scalable (Bi)CMOS manufacturing approach. To this end, both the gate stack and the Si/SiGe heterostructure must be of high quality, exhibiting uniformity across the wafer and consistent performance across multiple fabrication runs. Here, we report a comprehensive investigation of Si/SiGe heterostructures and gate stacks, fabricated in an industry-standard 200 mm BiCMOS pilot line. We evaluate the homogeneity and reproducibility by probing the properties of the two-dimensional electron gas (2DEG) in the shallow silicon quantum well through magnetotransport characterization of Hall bar-shaped field-effect transistors at 1.5 K. Across all the probed wafers, we observe minimal variation of the 2DEG properties, with an average maximum mobility of $(4.25\pm0.17)\times 10^{5}$ cm$^{2}$/Vs and low percolation carrier density of $(5.9\pm0.18)\times 10^{10}$ cm$^{-2}$ evidencing low disorder potential in the quantum well. The observed narrow statistical distribution of the transport properties highlights the reproducibility and the stability of the fabrication process. Furthermore, wafer-scale characterization of a selected individual wafer evidenced the homogeneity of the device performances across the wafer area. Based on these findings, we conclude that our material and processes provide a suitable platform for the development of scalable, Si/SiGe-based quantum devices. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:52:29 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Mistroni",
"Alberto",
""
],
[
"Lisker",
"Marco",
""
],
[
"Yamamoto",
"Yuji",
""
],
[
"Wen",
"Wei-Chen",
""
],
[
"Fidorra",
"Fabian",
""
],
[
"Tetzner",
"Henriette",
""
],
[
"Diebel",
"Laura K.",
""
],
[
"Visser",
"Lino",
""
],
[
"Anupam",
"Spandan",
""
],
[
"Mourik",
"Vincent",
""
],
[
"Schreiber",
"Lars R.",
""
],
[
"Bluhm",
"Hendrik",
""
],
[
"Bougeard",
"Dominique",
""
],
[
"Zoellner",
"Marvin H.",
""
],
[
"Capellini",
"Giovanni",
""
],
[
"Reichmann",
"Felix",
""
]
] |
2506.14661 | Nguyen Quang San | N. Q. San, O. D. Skoromnik, V. V. Triguk, I. D. Feranchuk | Approximating Hamiltonian for Hartree-Fock solutions for nonrelativistic atoms | null | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | In our work we construct a Hamiltonian, whose eigenstates approximate the solutions of the self-consistent Hartree-Fock equations for nonrelativistic atoms and ions. Its eigenvalues are given by completely algebraic expressions and the eigenfunctions are defined by Coulomb wave-functions orbitals. Within this approximation we compute the binding energy, ionization potentials, electron density distribution, electron density at the nucleus, and atomic scattering factors of nonrelativistic atoms and ions. The accuracy of our results is comparable with those obtained via the usage of the Hartree-Fock method but does not require solving integro-differential equations or numerically computing integrals with complex functions. This approach can serve as a good initial approximation for performing more accurate calculations and for the quantitative evaluation of physical parameters that depend on the electron density of atoms. The proposed approach is of interest for various fields of condensed matter physics, plasma physics and quantum chemistry. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:52:41 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"San",
"N. Q.",
""
],
[
"Skoromnik",
"O. D.",
""
],
[
"Triguk",
"V. V.",
""
],
[
"Feranchuk",
"I. D.",
""
]
] |
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"
}
] | 2025-06-18T00:00:00 | [
[
"Cho",
"Young-ho",
""
],
[
"Ko",
"Min-Seung",
""
],
[
"Zhu",
"Hao",
""
]
] |
2506.14663 | Chahra Rekaik | Chahra Rekaik and Mohamed Sadek Zidi | An extension of the SM with vector-like quarks and hypothetical heavy bosons: model independent parametrisation at NLO order | 50 pages | null | null | null | hep-ph | http://creativecommons.org/licenses/by/4.0/ | A model independent parameterization of an extension of the Standard Model including vector-like quarks, new heavy gauge bosons and an extra scalar, involving a minimal number of free parameters, is introduced. Theoretical constraints on the model couplings and hypothetical particle masses are established by making use of perturbative unitarity. The production of heavy charged and neutral gauge bosons which subsequently decay to vector-like quarks at the LHC are investigated at both leading-order and next-to-leading order. The predictions of the model are obtained by using 3 approaches: the narrow-width-approximation, the complex mass scheme and a mix of both, where the advantages and weaknesses of each method are emphasized. This study demonstrates that, for certain configurations, off-shell effects remain significant even when the width-to-mass ratios of the unstable particles are sufficiently small, rendering the narrow-width approximation unreliable and necessitating the use of the complex mass scheme. The predictions are compared with CMS {\tt run II} data, where lower bounds on the charged heavy gauge boson mass are set as function of the free parameter normalizing their coupling to quarks. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:53:58 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Rekaik",
"Chahra",
""
],
[
"Zidi",
"Mohamed Sadek",
""
]
] |
2506.14664 | Wolf-Peter Schill | Franziska Klaucke, Karsten Neuhoff, Alexander Roth, Wolf-Peter Schill, Leon Stolle | An advanced reliability reserve incentivizes flexibility investments while safeguarding the electricity market | null | null | null | null | econ.GN q-fin.EC | http://creativecommons.org/licenses/by/4.0/ | To ensure security of supply in the power sector, many countries are already using or discussing the introduction of capacity mechanisms. Two main types of such mechanisms include capacity markets and capacity reserves. Simultaneously, the expansion of variable renewable energy sources increases the need for power sector flexibility, for which there are promising yet often under-utilized options on the demand side. In this paper, we analyze how a centralized capacity market and an advanced reliability reserve with a moderately high activation price affect investments in demand-side flexibility technologies. We do so for a German case study of 2030, using an open-source capacity expansion model and incorporating detailed demand-side flexibility potentials across industry, process heat, and district heating. We show that a centralized capacity market effectively caps peak prices in the wholesale electricity market and thus reduces incentives for investments in demand-side flexibility options. The advanced reliability reserve induces substantially higher flexibility investments while leading to similar overall electricity supply costs and ensuring a similar level of security of supply. The advanced reliability reserve could thus create a learning environment for flexibility technologies to support the transition to climate neutral energy systems. Additionally, an advanced reliability reserve could be introduced faster and is more flexible than a centralized capacity market. While concrete design parameters are yet to be specified, we argue that policymakers should consider the reliability reserve concept in upcoming decision on capacity mechanisms in Germany and beyond. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:56:33 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Klaucke",
"Franziska",
""
],
[
"Neuhoff",
"Karsten",
""
],
[
"Roth",
"Alexander",
""
],
[
"Schill",
"Wolf-Peter",
""
],
[
"Stolle",
"Leon",
""
]
] |
2506.14666 | Sam P Fisher | Sam P. Fisher, Giovanni Italiano, Dawid Kielak | Virtual fibring of Poincar\'e-duality groups | 31 pages | null | null | null | math.GR | http://creativecommons.org/licenses/by/4.0/ | We show that a RFRS Poincar\'e-duality group $G$ admits a virtual epimorphism to the integers whose kernel is itself a Poincar\'e-duality group over every field if and only if the $L^2$-homology of $G$ vanishes and so do the positive-characteristic variants thereof.
Our investigations yield a more general relationship between cohomology at infinity of groups that algebraically fibre and their fibres. In particular, we show that if the fundamental group of an aspherical manifold of dimension at least three algebraically fibres, then the fibre is one ended. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 15:57:09 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Fisher",
"Sam P.",
""
],
[
"Italiano",
"Giovanni",
""
],
[
"Kielak",
"Dawid",
""
]
] |
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"
}
] | 2025-06-18T00:00:00 | [
[
"Poyser",
"Matt",
""
],
[
"Breckon",
"Toby P.",
""
]
] |
2506.14668 | Chengcai Shen | Chengcai Shen, Xiaocan Li, Yuan-Kuen Ko, John C. Raymond, Fan Guo, Vanessa Polito, and Viviane Pierrard | Modeling of Ionization and Recombination Processes in Plasma with Arbitrary Non-Maxwellian Electron Distributions | null | null | null | null | astro-ph.SR physics.plasm-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In astronomical environments, the high-temperature emission of plasma mainly depends on ion charge states, which requires accurate analysis of the ionization and recombination processes. For various phenomena involving energetic particles, the non-Maxwellian distributions of electrons exhibiting high-energy tails can significantly enhance the ionization process. Therefore, accurately computing ionization and recombination rates with non-Maxwellian electron distributions is essential for emission diagnostic analysis. In this work, we report two methods for fitting various non-Maxwellian distributions by using the Maxwellian decomposition strategy. For standard \{kappa} distributions, the calculated ionization and recombination rate coefficients show comparable accuracy to other public packages. We apply the above methods to two specific non-Maxwellian distribution scenarios: (I) accelerated electron distributions due to magnetic reconnection revealed in a combined MHD-particle simulation; (II) the high-energy truncated \{kappa} distribution predicted by the exospheric model of the solar wind. During the electron acceleration process, ionization rates of high-temperature iron ions increase significantly compared to their initial Maxwellian distribution, while the recombination rates may decrease due to the electron distribution changes in low-energy ranges. This can potentially lead to an overestimation of the plasma temperature when analyzing the Fe emission lines under the Maxwellian distribution assumption. For the truncated \{kappa} distribution in the solar wind, the ionization rates are lower than those for the standard \{kappa} distribution, while the recombination rates remain similar. This leads to an overestimation of plasma temperature when assuming a \{kappa} distribution. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:01:24 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Shen",
"Chengcai",
""
],
[
"Li",
"Xiaocan",
""
],
[
"Ko",
"Yuan-Kuen",
""
],
[
"Raymond",
"John C.",
""
],
[
"Guo",
"Fan",
""
],
[
"Polito",
"Vanessa",
""
],
[
"Pierrard",
"Viviane",
""
]
] |
2506.14669 | Federica Spoto | Federica Spoto, Jiazi Tian, Jonas H\"ugel, Daniel T. Ortega, Christine S. Ritchie, Deborah Blacker, Francesca Dominici, Chirag J. Patel, Daniel Mork, and Hossein Estiri | Quantifying Diagnostic Signal Decay in Dementia: A National Study of Medicare Hospitalization Data | null | null | null | null | stat.ME | http://creativecommons.org/licenses/by/4.0/ | Background: Artificial intelligence (AI) models in healthcare depend on the fidelity of diagnostic data, yet the quality of such data is often compromised by variability in clinical documentation practices. In dementia, a condition already prone to diagnostic ambiguity, this variability may introduce systematic distortion into claims-based research and AI model development.
Methods: We analyzed Medicare Part A hospitalization data from 2016-2018 to examine patterns of dementia-related ICD-10 code utilization across more than 3,000 U.S. counties. Using a clinically informed classification of 17 ICD-10 codes grouped into five diagnostic categories, we applied the transitive Sequential Pattern Mining (tSPM+) algorithm to model temporal usage structures. We then used matrix similarity methods to compare local diagnostic patterns to national norms and fit multivariable linear regressions to identify county-level demographic and structural correlates of divergence.
Findings: We found substantial geographic and demographic variation in dementia-related diagnostic code usage. Non-specific codes were dominant nationwide, while Alzheimer's disease and vascular dementia codes showed pronounced variability. Temporal sequence analysis revealed consistent transitions from specific to non-specific codes, which suggest degradation of diagnostic specificity over time. Counties with higher proportions of rural residents, Medicaid-eligible patients, and Black or Hispanic dementia patients demonstrated significantly lower similarity to national usage patterns. Our model explained 38% of the variation in local-to-national diagnostic alignment. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:02:11 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Spoto",
"Federica",
""
],
[
"Tian",
"Jiazi",
""
],
[
"Hügel",
"Jonas",
""
],
[
"Ortega",
"Daniel T.",
""
],
[
"Ritchie",
"Christine S.",
""
],
[
"Blacker",
"Deborah",
""
],
[
"Dominici",
"Francesca",
""
],
[
"Patel",
"Chirag J.",
""
],
[
"Mork",
"Daniel",
""
],
[
"Estiri",
"Hossein",
""
]
] |
2506.14670 | Jina Kim | Jina Kim, Leeje Jang, Yao-Yi Chiang, Guanyu Wang, Michelle Pasco | StreetLens: Enabling Human-Centered AI Agents for Neighborhood Assessment from Street View Imagery | null | null | null | null | cs.HC cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Traditionally, neighborhood studies have employed interviews, surveys, and manual image annotation guided by detailed protocols to identify environmental characteristics, including physical disorder, decay, street safety, and sociocultural symbols, and to examine their impact on developmental and health outcomes. While these methods yield rich insights, they are time-consuming and require intensive expert intervention. Recent technological advances, including vision-language models (VLMs), have begun to automate parts of this process; however, existing efforts are often ad hoc and lack adaptability across research designs and geographic contexts. In this demo paper, we present StreetLens, a human-centered, researcher-configurable workflow that embeds relevant social science expertise in a VLM for scalable neighborhood environmental assessments. StreetLens mimics the process of trained human coders by grounding the analysis in questions derived from established interview protocols, retrieving relevant street view imagery (SVI), and generating a wide spectrum of semantic annotations from objective features (e.g., the number of cars) to subjective perceptions (e.g., the sense of disorder in an image). By enabling researchers to define the VLM's role through domain-informed prompting, StreetLens places domain knowledge at the core of the analysis process. It also supports the integration of prior survey data to enhance robustness and expand the range of characteristics assessed across diverse settings. We provide a Google Colab notebook to make StreetLens accessible and extensible for researchers working with public or custom SVI datasets. StreetLens represents a shift toward flexible, agentic AI systems that work closely with researchers to accelerate and scale neighborhood studies. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:06:03 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Kim",
"Jina",
""
],
[
"Jang",
"Leeje",
""
],
[
"Chiang",
"Yao-Yi",
""
],
[
"Wang",
"Guanyu",
""
],
[
"Pasco",
"Michelle",
""
]
] |
2506.14671 | Yuji Odaka | Yuji Odaka | On Sun-Zhang's theory of Fano fibrations -- weighted volumes, moduli and bubbling Fano fibrations | v1: 33 pages | null | null | null | math.AG math.DG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We revisit the recent theory of Sun-Zhang on general Fano fibration (germs) which emerged from the study of non-compact Kahler-Ricci soliton metrics, primarily from an algebro-geometric perspective. In addition to reviewing the existing framework, we present new results, conjectures, and remarks. These include methods for computing weighted volumes via (restricted) volumes, Laplace transforms, and incomplete Gamma-functions, and a conjectural algebro-geometric construction (``bubbling") of asymptotically conical Fano fibrations from degenerating Fano fibrations. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:06:33 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Odaka",
"Yuji",
""
]
] |
2506.14672 | Chen Shen | Chen Shen, Honghu Zhang, Beate Kl\"osgen, Benjamin M. Ocko | Extending the capillary wave model to include the effect of bending rigidity: X-ray reflection and diffuse scattering | 30 pages, 11 figures | null | null | null | cond-mat.soft | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The surface roughness of a thin film at a liquid interface exhibits contributions of thermally excited fluctuations. This thermal roughness depends on temperature (T), surface tension ($\gamma$) and elastic material properties, specifically the bending modulus ($\kappa$) of the film. A non-zero $\kappa$ suppresses the thermal roughness at small length scales compared to an interface with zero $\kappa$, as expressed by the power spectral density (PSD) of the thermal roughness. The description of the X-ray scattering of the standard Capillary Wave Model (CWM), that is valid for zero $\kappa$, is extended to include the effect of $\kappa$. The extended CWM (eCWM) provides a single analytical form for both the specular XRR and the diffuse scattering around the specular reflection, and recovers the expression of the CWM at its zero $\kappa$ limit. This new theoretical approach enables the use of single-shot grazing incidence X-ray off-specular scattering (GIXOS) measurements for characterizing the structure of thin films on a liquid surface. The eCWM analysis approach decouples the thermal roughness factor from the surface scattering signal, providing direct access to the intrinsic surface-normal structure of the film and its bending modulus. Moreover, the eCWM facilitates the calculation of reflectivity at any desired resolution (pseudo XRR approach). The transformation into pseudo XRR provides the benefit of using widely available XRR software to perform GIXOS analysis. The extended range of the vertical scattering vector (Qz) available with the GIXOS-pseudo XRR approach allows for a higher spatial resolution than with conventional XRR. Experimental results are presented for various lipid systems, showing strong agreement between conventional specular XRR and pseudo XRR methods. This agreement validates the proposed approach and highlights its utility for analyzing soft, thin films. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:06:39 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Shen",
"Chen",
""
],
[
"Zhang",
"Honghu",
""
],
[
"Klösgen",
"Beate",
""
],
[
"Ocko",
"Benjamin M.",
""
]
] |
2506.14674 | Ling Li | Ling Li, Yao Zhou, Yuxuan Liang, Fugee Tsung, Jiaheng Wei | Recognition through Reasoning: Reinforcing Image Geo-localization with Large Vision-Language Models | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Previous methods for image geo-localization have typically treated the task as either classification or retrieval, often relying on black-box decisions that lack interpretability. The rise of large vision-language models (LVLMs) has enabled a rethinking of geo-localization as a reasoning-driven task grounded in visual cues. However, two major challenges persist. On the data side, existing reasoning-focused datasets are primarily based on street-view imagery, offering limited scene diversity and constrained viewpoints. On the modeling side, current approaches predominantly rely on supervised fine-tuning, which yields only marginal improvements in reasoning capabilities. To address these challenges, we propose a novel pipeline that constructs a reasoning-oriented geo-localization dataset, MP16-Reason, using diverse social media images. We introduce GLOBE, Group-relative policy optimization for Locatability assessment and Optimized visual-clue reasoning, yielding Bi-objective geo-Enhancement for the VLM in recognition and reasoning. GLOBE incorporates task-specific rewards that jointly enhance locatability assessment, visual clue reasoning, and geolocation accuracy. Both qualitative and quantitative results demonstrate that GLOBE outperforms state-of-the-art open-source LVLMs on geo-localization tasks, particularly in diverse visual scenes, while also generating more insightful and interpretable reasoning trajectories. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:07:58 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Li",
"Ling",
""
],
[
"Zhou",
"Yao",
""
],
[
"Liang",
"Yuxuan",
""
],
[
"Tsung",
"Fugee",
""
],
[
"Wei",
"Jiaheng",
""
]
] |
2506.14675 | Sihong Zhou | Jing Ou-Yang, Run-Hui Li, Si-Hong Zhou | Analysis of three-body charmed $B$ meson decays $B \to {D}(V^* \to){V P}$ | 25 pages, 2 figures | null | null | null | hep-ph | http://creativecommons.org/publicdomain/zero/1.0/ | We systematically analyze the decays $B_{(s)} \to D_{(s)} (V^* \to)\, V\, P$, where $V^*$ represents a vector resonance ($\rho, \, \omega$ or $K^*$), and $V P$ denotes the final-state meson pairs $ \omega\, \pi$, $ \rho\, \pi$ and $ \rho \, K$. The intermediate subprocesses $B_{(s)} \to D_{(s)} V^*$ are calculated in the factorization-assisted topological-amplitude approach, while the intermediate resonant states $V^*$ are modeled using a relativistic Breit-Wigner distribution, subsequently decaying into $VP$ through strong interactions. We predict the off-shell effects of the ground-state resonances ($\rho,\, \omega, \, K^*$) in $B_{(s)} \to D_{(s)} (V^* \to )V P$. Our results show that the virtual contributions from $\rho \to \omega \, \pi$, $\omega \to \rho\, \pi$, and $K^* \to \rho\, K$ are crucial for these three-body decays, $B_{(s)} \to D_{(s)} V\, P$. In particular, the branching fractions arising from the $\rho$ and $\omega$ virtual effects can be comparable to the total decay rates of $B_{(s)} \to D_{(s)} \omega\, \pi$ and $B_{(s)} \to D_{(s)} \rho\, \pi$, respectively. Decays with branching fractions of order $10^{-6}-10^{-4}$ are expected to be measurable at Belle II and LHCb. Compared with previous perturbative QCD predictions for $B_{(s)} \to D_{(s)} (\rho \to)\, \omega\, \pi$, our results are consistent but exhibit higher precision. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:08:04 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Ou-Yang",
"Jing",
""
],
[
"Li",
"Run-Hui",
""
],
[
"Zhou",
"Si-Hong",
""
]
] |
2506.14676 | Damien Querlioz | Mohammed Akib Iftakher, Hugo Levices, Kamel-Eddine Harabi, Adrien Renaudineau, Mathieu-Coumba Faye, Corentin Bouchard, Florian Disdier, Bernard Viala, Elisa Vianello, Philippe Talatchian, Kevin Garello, Damien Querlioz, Louis Hutin | Intrinsic Annealing in a Hybrid Memristor-Magnetic Tunnel Junction Ising Machine | null | null | null | null | cs.ET | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hardware implementations of the Ising model offer promising solutions to large-scale optimization tasks. In the literature, various nanodevices have been shown to emulate the spin dynamics for such Ising machines with remarkable effectiveness. Other nanodevices have been shown to implement spin-spin coupling with compact footprint and minimal energy dissipation. However, an ideal Ising machine would associate both types of nanodevices, and they must operate synergistically to support annealing: a progressive reduction of machine stochasticity that allows it to settle to energy minimum. Here, we report an Ising machine that combines two nanotechnologies: memristor crossbar -- storing multi-level couplings -- and stochastic magnetic tunnel junction (SMTJ), acting as thermally driven spins. Because the same read voltage that interrogates the crossbar also biases the SMTJs, increasing this voltage automatically lowers the effective temperature of the machine, providing an intrinsic, nearly circuit-free annealing technique. Operating at zero magnetic field, our prototype consistently reaches the global optimum of a 24-vertex weighted MAX-CUT and a 10-vertex, three-color graph-coloring problem. Given that both nanotechnologies in our demonstrator are CMOS-integrated, this approach is compatible with advanced 3D integration, offering a scalable pathway toward compact, fast, and energy-efficient large-scale Ising solvers. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:08:37 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Iftakher",
"Mohammed Akib",
""
],
[
"Levices",
"Hugo",
""
],
[
"Harabi",
"Kamel-Eddine",
""
],
[
"Renaudineau",
"Adrien",
""
],
[
"Faye",
"Mathieu-Coumba",
""
],
[
"Bouchard",
"Corentin",
""
],
[
"Disdier",
"Florian",
""
],
[
"Viala",
"Bernard",
""
],
[
"Vianello",
"Elisa",
""
],
[
"Talatchian",
"Philippe",
""
],
[
"Garello",
"Kevin",
""
],
[
"Querlioz",
"Damien",
""
],
[
"Hutin",
"Louis",
""
]
] |
2506.14677 | Yingchao Li | Yingchao Li | Design an Editable Speech-to-Sign-Language Transformer System: A Human-Centered AI Approach | null | null | null | null | cs.HC cs.AI | http://creativecommons.org/licenses/by/4.0/ | This paper presents a human-centered, real-time, user-adaptive speech-to-sign language animation system that integrates Transformer-based motion generation with a transparent, user-editable JSON intermediate layer. The framework overcomes key limitations in prior sign language technologies by enabling direct user inspection and modification of sign segments, thus enhancing naturalness, expressiveness, and user agency. Leveraging a streaming Conformer encoder and autoregressive Transformer-MDN decoder, the system synchronizes spoken input into upper-body and facial motion for 3D avatar rendering. Edits and user ratings feed into a human-in-the-loop optimization loop for continuous improvement. Experiments with 20 deaf signers and 5 interpreters show that the editable interface and participatory feedback significantly improve comprehension, naturalness, usability, and trust, while lowering cognitive load. With sub-20 ms per-frame inference on standard hardware, the system is ready for real-time communication and education. This work illustrates how technical and participatory innovation together enable accessible, explainable, and user-adaptive AI for sign language technology. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:08:48 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Li",
"Yingchao",
""
]
] |
2506.14678 | Isabella Mastroianni | Isabella Mastroianni, Marco Guerra, Ulderico Fugacci, Emanuela De Negri | Retrieving biparameter persistence modules from monoparameter ones: a characterization of hook-decomposable persistence modules | null | null | null | null | math.AT math.AC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Motivated by the need to relate the biparameter persistence module induced by a pair of scalar functions with the monoparameter persistence modules induced by each function separately, we introduce a construction that defines a kind of product between two monoparameter persistence modules. While originally conceived to serve this comparative purpose, our construction unexpectedly reveals a deeper structural property: it also characterizes a class of biparameter modules known as hook-decomposable modules. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:09:31 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Mastroianni",
"Isabella",
""
],
[
"Guerra",
"Marco",
""
],
[
"Fugacci",
"Ulderico",
""
],
[
"De Negri",
"Emanuela",
""
]
] |
2506.14679 | Catherine Gill | Michela Taufer, Rada Mihalcea, Matthew Turk, Dan Lopresti, Adam Wierman, Kevin Butler, Sven Koenig, David Danks, William Gropp, Manish Parashar, Yolanda Gil, Bill Regli, Rajmohan Rajaraman, David Jensen, Nadya Bliss, and Mary Lou Maher | Now More Than Ever, Foundational AI Research and Infrastructure Depends on the Federal Government | null | null | null | null | cs.CY | http://creativecommons.org/licenses/by/4.0/ | Leadership in the field of AI is vital for our nation's economy and security. Maintaining this leadership requires investments by the federal government. The federal investment in foundation AI research is essential for U.S. leadership in the field. Providing accessible AI infrastructure will benefit everyone. Now is the time to increase the federal support, which will be complementary to, and help drive, the nation's high-tech industry investments. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:09:53 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Taufer",
"Michela",
""
],
[
"Mihalcea",
"Rada",
""
],
[
"Turk",
"Matthew",
""
],
[
"Lopresti",
"Dan",
""
],
[
"Wierman",
"Adam",
""
],
[
"Butler",
"Kevin",
""
],
[
"Koenig",
"Sven",
""
],
[
"Danks",
"David",
""
],
[
"Gropp",
"William",
""
],
[
"Parashar",
"Manish",
""
],
[
"Gil",
"Yolanda",
""
],
[
"Regli",
"Bill",
""
],
[
"Rajaraman",
"Rajmohan",
""
],
[
"Jensen",
"David",
""
],
[
"Bliss",
"Nadya",
""
],
[
"Maher",
"Mary Lou",
""
]
] |
2506.14680 | Catherine Gill | Rada Mihalcea, Nazanin Andalibi, David Jensen, Matthew Turk, Pamela Wisniewski, and Holly Yanco | Which Humans? Inclusivity and Representation in Human-Centered AI | null | null | null | null | cs.CY | http://creativecommons.org/licenses/by/4.0/ | As AI systems continue to spread and become integrated into many aspects of society, the concept of "human-centered AI" has gained increasing prominence, raising the critical question of which humans are the AI systems to be centered around. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:12:43 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Mihalcea",
"Rada",
""
],
[
"Andalibi",
"Nazanin",
""
],
[
"Jensen",
"David",
""
],
[
"Turk",
"Matthew",
""
],
[
"Wisniewski",
"Pamela",
""
],
[
"Yanco",
"Holly",
""
]
] |
2506.14681 | Yuto Harada | Yuto Harada, Yusuke Yamauchi, Yusuke Oda, Yohei Oseki, Yusuke Miyao and Yu Takagi | Massive Supervised Fine-tuning Experiments Reveal How Data, Layer, and Training Factors Shape LLM Alignment Quality | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Supervised fine-tuning (SFT) is a critical step in aligning large language models (LLMs) with human instructions and values, yet many aspects of SFT remain poorly understood. We trained a wide range of base models on a variety of datasets including code generation, mathematical reasoning, and general-domain tasks, resulting in 1,000+ SFT models under controlled conditions. We then identified the dataset properties that matter most and examined the layer-wise modifications introduced by SFT. Our findings reveal that some training-task synergies persist across all models while others vary substantially, emphasizing the importance of model-specific strategies. Moreover, we demonstrate that perplexity consistently predicts SFT effectiveness--often surpassing superficial similarity between trained data and benchmark--and that mid-layer weight changes correlate most strongly with performance gains. We will release these 1,000+ SFT models and benchmark results to accelerate further research. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:13:15 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Harada",
"Yuto",
""
],
[
"Yamauchi",
"Yusuke",
""
],
[
"Oda",
"Yusuke",
""
],
[
"Oseki",
"Yohei",
""
],
[
"Miyao",
"Yusuke",
""
],
[
"Takagi",
"Yu",
""
]
] |
2506.14682 | Ads Dawson | Ads Dawson, Rob Mulla, Nick Landers, Shane Caldwell | AIRTBench: Measuring Autonomous AI Red Teaming Capabilities in Language Models | 43 pages, 13 figures, 16 tables | null | null | null | cs.CR | http://creativecommons.org/licenses/by-sa/4.0/ | We introduce AIRTBench, an AI red teaming benchmark for evaluating language models' ability to autonomously discover and exploit Artificial Intelligence and Machine Learning (AI/ML) security vulnerabilities. The benchmark consists of 70 realistic black-box capture-the-flag (CTF) challenges from the Crucible challenge environment on the Dreadnode platform, requiring models to write python code to interact with and compromise AI systems. Claude-3.7-Sonnet emerged as the clear leader, solving 43 challenges (61% of the total suite, 46.9% overall success rate), with Gemini-2.5-Pro following at 39 challenges (56%, 34.3% overall), GPT-4.5-Preview at 34 challenges (49%, 36.9% overall), and DeepSeek R1 at 29 challenges (41%, 26.9% overall). Our evaluations show frontier models excel at prompt injection attacks (averaging 49% success rates) but struggle with system exploitation and model inversion challenges (below 26%, even for the best performers). Frontier models are far outpacing open-source alternatives, with the best truly open-source model (Llama-4-17B) solving 7 challenges (10%, 1.0% overall), though demonstrating specialized capabilities on certain hard challenges. Compared to human security researchers, large language models (LLMs) solve challenges with remarkable efficiency completing in minutes what typically takes humans hours or days-with efficiency advantages of over 5,000x on hard challenges. Our contribution fills a critical gap in the evaluation landscape, providing the first comprehensive benchmark specifically designed to measure and track progress in autonomous AI red teaming capabilities. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:19:06 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Dawson",
"Ads",
""
],
[
"Mulla",
"Rob",
""
],
[
"Landers",
"Nick",
""
],
[
"Caldwell",
"Shane",
""
]
] |
2506.14683 | Yuntong Zhang | Leonhard Applis, Yuntong Zhang, Shanchao Liang, Nan Jiang, Lin Tan, Abhik Roychoudhury | Unified Software Engineering agent as AI Software Engineer | Leonhard Applis and Yuntong Zhang contributed equally to this work | null | null | null | cs.SE cs.AI | http://creativecommons.org/licenses/by/4.0/ | The growth of Large Language Model (LLM) technology has raised expectations for automated coding. However, software engineering is more than coding and is concerned with activities including maintenance and evolution of a project. In this context, the concept of LLM agents has gained traction, which utilize LLMs as reasoning engines to invoke external tools autonomously. But is an LLM agent the same as an AI software engineer? In this paper, we seek to understand this question by developing a Unified Software Engineering agent or USEagent. Unlike existing work which builds specialized agents for specific software tasks such as testing, debugging, and repair, our goal is to build a unified agent which can orchestrate and handle multiple capabilities. This gives the agent the promise of handling complex scenarios in software development such as fixing an incomplete patch, adding new features, or taking over code written by others. We envision USEagent as the first draft of a future AI Software Engineer which can be a team member in future software development teams involving both AI and humans. To evaluate the efficacy of USEagent, we build a Unified Software Engineering bench (USEbench) comprising of myriad tasks such as coding, testing, and patching. USEbench is a judicious mixture of tasks from existing benchmarks such as SWE-bench, SWT-bench, and REPOCOD. In an evaluation on USEbench consisting of 1,271 repository-level software engineering tasks, USEagent shows improved efficacy compared to existing general agents such as OpenHands CodeActAgent. There exist gaps in the capabilities of USEagent for certain coding tasks, which provides hints on further developing the AI Software Engineer of the future. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:19:13 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Applis",
"Leonhard",
""
],
[
"Zhang",
"Yuntong",
""
],
[
"Liang",
"Shanchao",
""
],
[
"Jiang",
"Nan",
""
],
[
"Tan",
"Lin",
""
],
[
"Roychoudhury",
"Abhik",
""
]
] |
2506.14684 | Aditya Bhattacharjee | Aditya Bhattacharjee, Ivan Meresman Higgs, Mark Sandler, Emmanouil Benetos | Refining music sample identification with a self-supervised graph neural network | Accepted at International Conference for Music Information Retrieval (ISMIR) 2025 | null | null | null | cs.SD cs.AI cs.IR | http://creativecommons.org/licenses/by/4.0/ | Automatic sample identification (ASID), the detection and identification of portions of audio recordings that have been reused in new musical works, is an essential but challenging task in the field of audio query-based retrieval. While a related task, audio fingerprinting, has made significant progress in accurately retrieving musical content under "real world" (noisy, reverberant) conditions, ASID systems struggle to identify samples that have undergone musical modifications. Thus, a system robust to common music production transformations such as time-stretching, pitch-shifting, effects processing, and underlying or overlaying music is an important open challenge.
In this work, we propose a lightweight and scalable encoding architecture employing a Graph Neural Network within a contrastive learning framework. Our model uses only 9% of the trainable parameters compared to the current state-of-the-art system while achieving comparable performance, reaching a mean average precision (mAP) of 44.2%.
To enhance retrieval quality, we introduce a two-stage approach consisting of an initial coarse similarity search for candidate selection, followed by a cross-attention classifier that rejects irrelevant matches and refines the ranking of retrieved candidates - an essential capability absent in prior models. In addition, because queries in real-world applications are often short in duration, we benchmark our system for short queries using new fine-grained annotations for the Sample100 dataset, which we publish as part of this work. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:19:21 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Bhattacharjee",
"Aditya",
""
],
[
"Higgs",
"Ivan Meresman",
""
],
[
"Sandler",
"Mark",
""
],
[
"Benetos",
"Emmanouil",
""
]
] |
2506.14685 | Erik Burman | Erik Burman and Mingfei Lu | Posterior contraction rates of computational methods for Bayesian data assimilation | null | null | null | null | math.NA cs.NA math.PR math.ST stat.TH | http://creativecommons.org/licenses/by/4.0/ | In this paper, we analyze posterior consistency of a Bayesian data assimilation problem under discretization. We prove convergence rates for the discrete posterior to ground truth solution under both conforming discretization and finite element discretization (usually non-conforming). The analysis is based on the coupling of asymptotics between the number of samples and the dimension of discrete spaces. In the finite element discretization, tailor-made discrete priors, instead of the discretization of continuous priors, are used to generate an optimal convergence rate. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:20:07 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Burman",
"Erik",
""
],
[
"Lu",
"Mingfei",
""
]
] |
2506.14686 | Xi Chen | Xi Chen, Hengshuang Zhao | FocalClick-XL: Towards Unified and High-quality Interactive Segmentation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Interactive segmentation enables users to extract binary masks of target objects through simple interactions such as clicks, scribbles, and boxes. However, existing methods often support only limited interaction forms and struggle to capture fine details. In this paper, we revisit the classical coarse-to-fine design of FocalClick and introduce significant extensions. Inspired by its multi-stage strategy, we propose a novel pipeline, FocalClick-XL, to address these challenges simultaneously. Following the emerging trend of large-scale pretraining, we decompose interactive segmentation into meta-tasks that capture different levels of information -- context, object, and detail -- assigning a dedicated subnet to each level.This decomposition allows each subnet to undergo scaled pretraining with independent data and supervision, maximizing its effectiveness. To enhance flexibility, we share context- and detail-level information across different interaction forms as common knowledge while introducing a prompting layer at the object level to encode specific interaction types. As a result, FocalClick-XL achieves state-of-the-art performance on click-based benchmarks and demonstrates remarkable adaptability to diverse interaction formats, including boxes, scribbles, and coarse masks. Beyond binary mask generation, it is also capable of predicting alpha mattes with fine-grained details, making it a versatile and powerful tool for interactive segmentation. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:21:32 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Chen",
"Xi",
""
],
[
"Zhao",
"Hengshuang",
""
]
] |
2506.14687 | Johnpierre Paglione | Ram Kumar, Shanta R. Saha, Jarryd Horn, A. Ikeda, Danila Sokratov, Yash Anand, Prathum Saraf, Ryan Dorman, E. Hemley, K. K. Iyer, Johnpierre Paglione | Complex single-site magnetism and magnetotransport in single-crystalline Gd$_{2}$AlSi$_{3}$ | 10 pages, 9 figures | Phys. Rev. B 111, 214426 (2025) | 10.1103/1sc3-6cfp | null | cond-mat.str-el cond-mat.mtrl-sci | http://creativecommons.org/licenses/by/4.0/ | We present a detailed investigation of single-crystal samples of the magnetic compound Gd$_{2}$AlSi$_{3}$, which crystallizes in the $\alpha$-ThSi$_2$ type tetragonal structure. We report the temperature and magnetic field dependence of the magnetic susceptibility, magnetization, heat capacity, electrical resistivity, and magnetoresistance for magnetic fields applied along both the tetragonal $c$-axis and in the basal $ab$-plane. X-ray diffraction measurements confirm a centrosymmetric, $I4_{1}/amd$ space group of the crystal structure. Despite single-site occupancy of the Gd position in this tetragonal structure, we identify two successive antiferromagnetic phase transitions at Ne\'el temperatures 32~K and 23~K via magnetic susceptibility, heat capacity and transport measurements, as well as a complex magnetic interaction with a magnetic anisotropy that plays an important role in the direction-dependent transport response. Our identification of multiple magnetic phases in Gd$_{2}$AlSi$_{3}$, where Gd is the only magnetic species, helps to elucidate the field-induced skyrmionic behavior in the Gd-based intermetallic compounds. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:23:30 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Kumar",
"Ram",
""
],
[
"Saha",
"Shanta R.",
""
],
[
"Horn",
"Jarryd",
""
],
[
"Ikeda",
"A.",
""
],
[
"Sokratov",
"Danila",
""
],
[
"Anand",
"Yash",
""
],
[
"Saraf",
"Prathum",
""
],
[
"Dorman",
"Ryan",
""
],
[
"Hemley",
"E.",
""
],
[
"Iyer",
"K. K.",
""
],
[
"Paglione",
"Johnpierre",
""
]
] |
2506.14688 | Andrew Potter | Shival Dasu, Simon Burton, Karl Mayer, David Amaro, Justin A. Gerber, Kevin Gilmore, Dan Gresh, Davide DelVento, Andrew C. Potter, David Hayes | Breaking even with magic: demonstration of a high-fidelity logical non-Clifford gate | 7+5 pages, 8 figures | null | null | null | quant-ph | http://creativecommons.org/licenses/by/4.0/ | Encoding quantum information to protect it from errors is essential for performing large-scale quantum computations. Performing a universal set of quantum gates on encoded states demands a potentially large resource overhead and minimizing this overhead is key for the practical development of large-scale fault-tolerant quantum computers. We propose and experimentally implement a magic-state preparation protocol to fault-tolerantly prepare a pair of logical magic states in a [[6,2,2]] quantum error-detecting code using only eight physical qubits. Implementing this protocol on H1-1, a 20 qubit trapped-ion quantum processor, we prepare magic states with experimental infidelity $7^{+3}_{-1}\times 10^{-5}$ with a $14.8^{+1}_{-1}\%$ discard rate and use these to perform a fault-tolerant non-Clifford gate, the controlled-Hadamard (CH), with logical infidelity $\leq 2.3^{+9}_{-9}\times 10^{-4}$. Notably, this significantly outperforms the unencoded physical CH infidelity of $10^{-3}$. Through circuit-level stabilizer simulations, we show that this protocol can be self-concatenated to produce extremely high-fidelity magic states with low space-time overhead in a [[36,4,4]] quantum error correcting code, with logical error rates of $6\times 10^{-10}$ ($5\times 10^{-14}$) at two-qubit error rate of $10^{-3}$ ($10^{-4}$) respectively. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:23:47 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Dasu",
"Shival",
""
],
[
"Burton",
"Simon",
""
],
[
"Mayer",
"Karl",
""
],
[
"Amaro",
"David",
""
],
[
"Gerber",
"Justin A.",
""
],
[
"Gilmore",
"Kevin",
""
],
[
"Gresh",
"Dan",
""
],
[
"DelVento",
"Davide",
""
],
[
"Potter",
"Andrew C.",
""
],
[
"Hayes",
"David",
""
]
] |
2506.14689 | So Nakamura | So Nakamura | A ringed-space-like structure on coalgebras for noncommutative algebraic geometry | 48 pages, comments welcome | null | null | null | math.RA math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Inspired by the perspective of Reyes' noncomutative spectral theory, we attempt to develop noncommutative algebraic geometry by introducing ringed coalgebras, which can be thought of as a noncommutative generalization of schemes over a field $k$. These objects arise from fully residually finite-dimensional(RFD) algebras introduced by Reyes and from schemes locally of finite type over $k$. The construction uses the Heyneman-Sweedler finite dual coalgebra and the Takeuchi underlying coalgebra. When $k$ is algebraically closed, the formation of ringed coalgebras gives a fully faithful functor out of the category of fully RFD algebras, as well as a fully faithful functor out of the category of schemes locally of finite type. The restrictions of these two functors to the category of (commutative) finitely generated algebras are isomorphic. Finally, we introduce modules over ringed coalgebras and show that the category of finitely generated modules on a fully RFD algebra and that of coherent sheaves on a scheme locally of finite type can, if $k$ is algebraically closed, be fully faithfully embedded into the category of modules over the corresponding ringed coalgebras. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:24:42 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Nakamura",
"So",
""
]
] |
2506.14690 | Kalliyan Velasco | Kalliyan Velasco, Timothy W. McLain, and Joshua G. Mangelson | Factor-Graph-Based Passive Acoustic Navigation for Decentralized Cooperative Localization Using Bearing Elevation Depth Difference | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Accurate and scalable underwater multi-agent localization remains a critical challenge due to the constraints of underwater communication. In this work, we propose a multi-agent localization framework using a factor-graph representation that incorporates bearing, elevation, and depth difference (BEDD). Our method leverages inverted ultra-short baseline (inverted-USBL) derived azimuth and elevation measurements from incoming acoustic signals and relative depth measurements to enable cooperative localization for a multi-robot team of autonomous underwater vehicles (AUVs). We validate our approach in the HoloOcean underwater simulator with a fleet of AUVs, demonstrating improved localization accuracy compared to dead reckoning. Additionally, we investigate the impact of azimuth and elevation measurement outliers, highlighting the need for robust outlier rejection techniques for acoustic signals. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:28:51 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Velasco",
"Kalliyan",
""
],
[
"McLain",
"Timothy W.",
""
],
[
"Mangelson",
"Joshua G.",
""
]
] |
2506.14691 | Zachary Garvin | Zachary K. Garvin (1), Ana\"is Roussel (1), Luoth Chou (2), Marco E. Castillo (2 and 3), Xiang Li (2), William B. Brinckerhoff (2), Sarah Stewart Johnson (1) ((1) Georgetown University, Washington, D.C., USA, (2) NASA Goddard Space Flight Center, Greenbelt, MD, USA, (3) Aerodyne Industries, Cape Canaveral, FL, USA) | Application of a modified commercial laser mass spectrometer as a science analog of the Mars Organic Molecule Analyzer (MOMA) | Submitted to Astrobiology. ZKG and AR contributed equally to this work. 24 pages, 11 figures, 3 tables (including Supplemental) | null | null | null | astro-ph.EP astro-ph.IM | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The ESA/NASA Rosalind Franklin rover, planned for launch in 2028, will carry the first laser desorption ionization mass spectrometer (LDI-MS) to Mars as part of the Mars Organic Molecule Analyzer (MOMA) instrument. MOMA will contribute to the astrobiology goals of the mission through the analysis of potential organic biosignatures. Due to the minimal availability of comparable equipment, laboratory analyses using similar techniques and instrumentation have been limited. In this study, we present a modified commercial benchtop LDI-MS designed to replicate MOMA functionality and to enable rapid testing of samples for MOMA validation experiments. We demonstrate that our instrument can detect organic standards in mineral matrices, with MS/MS enabling structural identification even in complex mixtures. Performance was additionally validated against an existing LDI-MS prototype through the comparison of spectra derived from natural samples from a Mars analog site in the Atacama Desert. Lastly, analysis of Mars analog synthetic mineral mixes highlights the capacity of the instrument to characterize both the mineralogical and organic signals in mission-relevant samples. This modified benchtop instrument will serve as a platform for collaborative research to prepare for MOMA operations, test LDI parameters, and generate pre-flight reference data in support of the mission science and astrobiology specific goals. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:29:09 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Garvin",
"Zachary K.",
"",
"Georgetown University, Washington, D.C., USA"
],
[
"Roussel",
"Anaïs",
"",
"Georgetown University, Washington, D.C., USA"
],
[
"Chou",
"Luoth",
"",
"NASA Goddard Space Flight Center, Greenbelt, MD, USA"
],
[
"Castillo",
"Marco E.",
"",
"2 and 3"
],
[
"Li",
"Xiang",
"",
"NASA Goddard Space Flight Center, Greenbelt, MD, USA"
],
[
"Brinckerhoff",
"William B.",
"",
"NASA Goddard Space Flight Center, Greenbelt, MD, USA"
],
[
"Johnson",
"Sarah Stewart",
"",
"Georgetown University, Washington, D.C., USA"
]
] |
2506.14692 | Chiara D'ercoli | Chiara D'Ercoli, Giulia Di Teodoro, Federico Siciliano | A Systematic Replicability and Comparative Study of BSARec and SASRec for Sequential Recommendation | null | null | null | null | cs.IR | http://creativecommons.org/licenses/by/4.0/ | This study aims at comparing two sequential recommender systems: Self-Attention based Sequential Recommendation (SASRec), and Beyond Self-Attention based Sequential Recommendation (BSARec) in order to check the improvement frequency enhancement - the added element in BSARec - has on recommendations. The models in the study, have been re-implemented with a common base-structure from EasyRec, with the aim of obtaining a fair and reproducible comparison. The results obtained displayed how BSARec, by including bias terms for frequency enhancement, does indeed outperform SASRec, although the increases in performance obtained, are not as high as those presented by the authors. This work aims at offering an overview on existing methods, and most importantly at underlying the importance of implementation details for performance comparison. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:29:55 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"D'Ercoli",
"Chiara",
""
],
[
"Di Teodoro",
"Giulia",
""
],
[
"Siciliano",
"Federico",
""
]
] |
2506.14693 | Samuel Fedida | Samuel Fedida | Einstein causality of quantum measurements in the Tomonaga-Schwinger picture | 17 pages, 4 figures | null | null | null | quant-ph gr-qc hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | When quantum measurements are conducted over spacelike-separated regions of spacetime, a natural and commonly assumed physical postulate, called Einstein causality, asserts that they should commute. In this paper, we provide a generalisation to L\"uders' rule \`a la Aharonov-Albert in those globally hyperbolic spacetimes which allow unitarily equivalent Hilbert spaces to be defined along Cauchy hypersurfaces, thus relying on the existence of an interaction picture \`a la Tomonaga-Schwinger. We show that under this rule, selective quantum measurements satisfy a state-independent anyonic commutation relation over spacelike-separated (pre)compact regions. We highlight that this propagates to positive operator-valued measures (POVMs), where the commutation is necessarily bosonic. We then examine Sorkin's impossible measurements and show that immediate contradictions can be averted as long as collapse-inducing measurements are irreversible. We finish by discussing the possibility of extending such results beyond the interaction picture. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:32:29 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Fedida",
"Samuel",
""
]
] |
2506.14694 | Andr\'as M\'esz\'aros | Andr\'as M\'esz\'aros | The homology torsion growth of determinantal hypertrees | null | null | null | null | math.CO math.AT math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Fix a dimension $d\ge 2$, and let $T_n$ be a random $d$-dimensional determinantal hypertree on $n$ vertices. We prove that \[\frac{\log|H_{d-1}(T_n,\mathbb{Z})|}{{{n\choose {d}}}}\] converges in probability to a constant $c_d$, which satisfies
\[\frac{1}2 \log\left(\frac{d+1}e\right)\le c_d\le \frac{1}2 \log\left(d+1\right) .\] | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:32:41 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Mészáros",
"András",
""
]
] |
2506.14695 | Adrian Ka-Wai Chung | Adrian Ka-Wai Chung, Nicol\'as Yunes | Probing quadratic gravity with black-hole ringdown gravitational waves measured by LIGO-Virgo-KAGRA detectors | 5 pages of main text, 6 pages of appendices | null | null | null | gr-qc astro-ph.HE | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quadratic gravity theories emerge as the low-energy limit of many grand-unified and quantum-gravity theories. We report the first gravitational-wave ringdown constraints on such theories obtained purely from the quasinormal-mode spectra of rapidly-spinning black-hole remnants. We find no quadratic-gravity signatures from ringdown signals measured by the LIGO-Virgo-KAGRA detectors, thereby constraining the coupling length scale of axi-dilaton gravity below 34 km, dynamical Chern-Simons gravity below 49 km, and scalar Gauss-Bonnet gravity below 47 km. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:33:47 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Chung",
"Adrian Ka-Wai",
""
],
[
"Yunes",
"Nicolás",
""
]
] |
2506.14697 | Zonghao Ying | Aishan Liu, Zonghao Ying, Le Wang, Junjie Mu, Jinyang Guo, Jiakai Wang, Yuqing Ma, Siyuan Liang, Mingchuan Zhang, Xianglong Liu, Dacheng Tao | AGENTSAFE: Benchmarking the Safety of Embodied Agents on Hazardous Instructions | 11 pages | null | null | null | cs.CR cs.RO | http://creativecommons.org/licenses/by/4.0/ | The rapid advancement of vision-language models (VLMs) and their integration into embodied agents have unlocked powerful capabilities for decision-making. However, as these systems are increasingly deployed in real-world environments, they face mounting safety concerns, particularly when responding to hazardous instructions. In this work, we propose AGENTSAFE, the first comprehensive benchmark for evaluating the safety of embodied VLM agents under hazardous instructions. AGENTSAFE simulates realistic agent-environment interactions within a simulation sandbox and incorporates a novel adapter module that bridges the gap between high-level VLM outputs and low-level embodied controls. Specifically, it maps recognized visual entities to manipulable objects and translates abstract planning into executable atomic actions in the environment. Building on this, we construct a risk-aware instruction dataset inspired by Asimovs Three Laws of Robotics, including base risky instructions and mutated jailbroken instructions. The benchmark includes 45 adversarial scenarios, 1,350 hazardous tasks, and 8,100 hazardous instructions, enabling systematic testing under adversarial conditions ranging from perception, planning, and action execution stages. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:37:35 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Liu",
"Aishan",
""
],
[
"Ying",
"Zonghao",
""
],
[
"Wang",
"Le",
""
],
[
"Mu",
"Junjie",
""
],
[
"Guo",
"Jinyang",
""
],
[
"Wang",
"Jiakai",
""
],
[
"Ma",
"Yuqing",
""
],
[
"Liang",
"Siyuan",
""
],
[
"Zhang",
"Mingchuan",
""
],
[
"Liu",
"Xianglong",
""
],
[
"Tao",
"Dacheng",
""
]
] |
2506.14698 | Sidney Bender | Sidney Bender, Jan Herrmann, Klaus-Robert M\"uller, Gr\'egoire Montavon | Towards Desiderata-Driven Design of Visual Counterfactual Explainers | null | null | null | null | cs.LG cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Visual counterfactual explainers (VCEs) are a straightforward and promising approach to enhancing the transparency of image classifiers. VCEs complement other types of explanations, such as feature attribution, by revealing the specific data transformations to which a machine learning model responds most strongly. In this paper, we argue that existing VCEs focus too narrowly on optimizing sample quality or change minimality; they fail to consider the more holistic desiderata for an explanation, such as fidelity, understandability, and sufficiency. To address this shortcoming, we explore new mechanisms for counterfactual generation and investigate how they can help fulfill these desiderata. We combine these mechanisms into a novel 'smooth counterfactual explorer' (SCE) algorithm and demonstrate its effectiveness through systematic evaluations on synthetic and real data. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:38:15 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Bender",
"Sidney",
""
],
[
"Herrmann",
"Jan",
""
],
[
"Müller",
"Klaus-Robert",
""
],
[
"Montavon",
"Grégoire",
""
]
] |
2506.14699 | Zimo Sun | Simone Giombi and Zimo Sun | Higher loops in AdS: applications to boundary CFT | 34 pages, 4 figures, 2 tables | null | null | null | hep-th | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The Euclidean Anti-de Sitter (AdS) space provides a natural framework for studying boundary conformal field theory (BCFT). We analyze the conformal boundary conditions of the critical O$(N)$ model in $d=4-\epsilon$ dimensions using the $\epsilon$-expansion, and extract some BCFT observables through higher-loop calculations in AdS. Specifically, in the so-called "ordinary" universality class, we determine the free energy to four-loop order and the one-point function of the lightest O$(N)$ singlet operator to three-loop order. In the symmetry breaking "normal" universality class, we derive the two-loop free energy and compute the leading correction to the one-point function of the lightest O($N$) vector. We apply Pad\'e approximants to extract the corresponding conformal data in three dimensions. In particular, from a suitable dimensional continuation of the free energy in AdS, we obtain estimates for the boundary central charge of the BCFT in $d=3$. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:38:17 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Giombi",
"Simone",
""
],
[
"Sun",
"Zimo",
""
]
] |
2506.14700 | ATLAS Collaboration | ATLAS Collaboration | Measurements of differential cross-sections of $WbWb$ production in the dilepton channel in $pp$ collisions at $\sqrt{s}$ = 13 TeV using the ATLAS detector | 64 pages in total, author list starting page 47, 14 figures, 6 tables, submitted to JHEP. All figures including auxiliary figures are available at http://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/TOPQ-2020-14 | null | null | CERN-EP-2025-094 | hep-ex | http://creativecommons.org/licenses/by/4.0/ | At the Large Hadron Collider, the $WbWb$ final state is expected to be dominated by $t\bar{t}$ production with a contribution from single-top processes. Differential cross-sections for $WbWb$ production in the dilepton decay channel are measured at the particle level as a function of various kinematic variables. The analysis is based on data from proton-proton collisions at a centre-of-mass energy of $\sqrt{s} = 13$ TeV, recorded by the ATLAS detector at the Large Hadron Collider over the period from 2015 to 2018, corresponding to an integrated luminosity of 140 fb$^{-1}$. Measurements are performed within the fiducial phase-space defined by the presence of two $b$-jets and one electron and one muon of opposite charges. The differential cross-sections are corrected for detector effects and unfolded to the particle level. Results are compared with predictions from Monte Carlo event generators at next-to-leading order in perturbative quantum chromodynamics. These measurements provide valuable constraints on the modelling of $WbWb$ production and the interference between doubly resonant and singly resonant $WbWb$ production. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:38:52 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"ATLAS Collaboration",
"",
""
]
] |
2506.14701 | Mingzhe Hu | Chih-Wei Chang, Sri Akkineni, Mingzhe Hu, Keyur D. Shah, Jun Zhou, Xiaofeng Yang | A Digital Twin Framework for Adaptive Treatment Planning in Radiotherapy | null | null | null | null | physics.med-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This study aims to develop and evaluate a digital twin (DT) framework to enhance adaptive proton therapy for prostate stereotactic body radiotherapy (SBRT), focusing on improving treatment precision for dominant intraprostatic lesions (DILs) while minimizing organ-at-risk (OAR) toxicity. We propose a decision-theoretic (DT) framework combining deep learning (DL)-based deformable image registration (DIR) with a prior treatment database to generate synthetic CTs (sCTs) for predicting interfractional anatomical changes. Using daily CBCT from five prostate SBRT patients with DILs, the framework precomputes multiple plans with high (DT-H) and low (DT-L) similarity sCTs. Plan optimization is performed in RayStation 2023B, assuming a constant RBE of 1.1 and robustly accounting for positional and range uncertainties. Plan quality is evaluated via a modified ProKnow score across two fractions, with reoptimization limited to 10 minutes. Daily CBCT evaluation showed clinical plans often violated OAR constraints (e.g., bladder V20.8Gy, rectum V23Gy), with DIL V100 < 90% in 2 patients, indicating SIFB failure. DT-H plans, using high-similarity sCTs, achieved better or comparable DIL/CTV coverage and lower OAR doses, with reoptimization completed within 10 min (e.g., DT-H-REopt-A score: 154.3-165.9). DT-L plans showed variable outcomes; lower similarity correlated with reduced DIL coverage (e.g., Patient 4: 84.7%). DT-H consistently outperformed clinical plans within time limits, while extended optimization brought DT-L and clinical plans closer to DT-H quality. This DT framework enables rapid, personalized adaptive proton therapy, improving DIL targeting and reducing toxicity. By addressing geometric uncertainties, it supports outcome gains in ultra-hypofractionated prostate RT and lays groundwork for future multimodal anatomical prediction. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:40:10 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Chang",
"Chih-Wei",
""
],
[
"Akkineni",
"Sri",
""
],
[
"Hu",
"Mingzhe",
""
],
[
"Shah",
"Keyur D.",
""
],
[
"Zhou",
"Jun",
""
],
[
"Yang",
"Xiaofeng",
""
]
] |
2506.14702 | Daniel D'souza | Daniel D'souza, Julia Kreutzer, Adrien Morisot, Ahmet \"Ust\"un, Sara Hooker | Treasure Hunt: Real-time Targeting of the Long Tail using Training-Time Markers | null | null | null | null | cs.CL cs.LG | http://creativecommons.org/licenses/by-nc-nd/4.0/ | One of the most profound challenges of modern machine learning is performing well on the long-tail of rare and underrepresented features. Large general-purpose models are trained for many tasks, but work best on high-frequency use cases. After training, it is hard to adapt a model to perform well on specific use cases underrepresented in the training corpus. Relying on prompt engineering or few-shot examples to maximize the output quality on a particular test case can be frustrating, as models can be highly sensitive to small changes, react in unpredicted ways or rely on a fixed system prompt for maintaining performance. In this work, we ask: "Can we optimize our training protocols to both improve controllability and performance on underrepresented use cases at inference time?" We revisit the divide between training and inference techniques to improve long-tail performance while providing users with a set of control levers the model is trained to be responsive to. We create a detailed taxonomy of data characteristics and task provenance to explicitly control generation attributes and implicitly condition generations at inference time. We fine-tune a base model to infer these markers automatically, which makes them optional at inference time. This principled and flexible approach yields pronounced improvements in performance, especially on examples from the long tail of the training distribution. While we observe an average lift of 5.7% win rates in open-ended generation quality with our markers, we see over 9.1% gains in underrepresented domains. We also observe relative lifts of up to 14.1% on underrepresented tasks like CodeRepair and absolute improvements of 35.3% on length instruction following evaluations. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:40:42 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"D'souza",
"Daniel",
""
],
[
"Kreutzer",
"Julia",
""
],
[
"Morisot",
"Adrien",
""
],
[
"Üstün",
"Ahmet",
""
],
[
"Hooker",
"Sara",
""
]
] |
2506.14703 | Stijn Vincent Kleijweg | Stijn V. Kleijweg and Philippe Corboz | Zigzag antiferromagnets in the SU(3) Hubbard model on the square lattice | 10 pages, 7 figures | null | null | null | cond-mat.str-el cond-mat.quant-gas | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | SU(N) Hubbard models exhibit a rich variety of phases, which may be realized through quantum simulation with ultracold atomic gases in optical lattices. In this work we study the Mott insulating phases of the SU(3) Hubbard model at 1/3-filling using infinite projected entangled-pair states, optimized with both imaginary time evolution and variational optimization. In the limit of strong interactions we reproduce the antiferromagnetic 3-sublattice ordered state previously identified in the SU(3) Heisenberg model. At intermediate interaction strength we find antiferromagnetic states exhibiting zigzag patterns of different lengths, in agreement with previous Hartree-Fock and constrained-path auxiliary-field quantum Monte Carlo calculations. We study the color order parameter and energy anisotropy, which are discontinuous across the phase transitions. Finally, we analyze the different energy contributions in two competing phases, identifying low-energy bonds at the corners of the zigzag that help stabilize the zigzag states. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:41:01 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Kleijweg",
"Stijn V.",
""
],
[
"Corboz",
"Philippe",
""
]
] |
2506.14704 | Anton Changalidis | Anton Changalidis and Aki H\"arm\"a | Capacity Matters: a Proof-of-Concept for Transformer Memorization on Real-World Data | This work has been accepted for publication at the First Workshop on Large Language Model Memorization (L2M2) at ACL 2025, Vienna, Austria | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | This paper studies how the model architecture and data configurations influence the empirical memorization capacity of generative transformers. The models are trained using synthetic text datasets derived from the Systematized Nomenclature of Medicine (SNOMED) knowledge graph: triplets, representing static connections, and sequences, simulating complex relation patterns. The results show that embedding size is the primary determinant of learning speed and capacity, while additional layers provide limited benefits and may hinder performance on simpler datasets. Activation functions play a crucial role, and Softmax demonstrates greater stability and capacity. Furthermore, increasing the complexity of the data set seems to improve the final memorization. These insights improve our understanding of transformer memory mechanisms and provide a framework for optimizing model design with structured real-world data. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:42:54 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Changalidis",
"Anton",
""
],
[
"Härmä",
"Aki",
""
]
] |
2506.14705 | Martin Bojowald | Joseph Balsells and Martin Bojowald | Relativistic implications of entropy and purity | 11 pages, one figure | null | null | null | quant-ph gr-qc | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A quantum object is extended by virtue of quantum uncertainty. Subjected to gravity, it therefore experiences tidal forces and other relativistic effects. As a result, entropy and purity affect geodesic motion and acquire weight, an observation that has broad implications from free-fall experiments to Hawking radiation. If the object's position is quantum correlated in at least two directions, a complete description of its geodesic motion requires non-Riemannian geometry of a form controlled by the entropy and purity of its state. In addition, the analysis reveals a new Minkowski structure on a subspace of entanglement degrees of freedom. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:43:11 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Balsells",
"Joseph",
""
],
[
"Bojowald",
"Martin",
""
]
] |
2506.14706 | Ni Ou | Ni Ou, Zhuo Chen, Xinru Zhang and Junzheng Wang | Iterative Camera-LiDAR Extrinsic Optimization via Surrogate Diffusion | 7 pages, 4 figures, accepted by IROS 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Cameras and LiDAR are essential sensors for autonomous vehicles. The fusion of camera and LiDAR data addresses the limitations of individual sensors but relies on precise extrinsic calibration. Recently, numerous end-to-end calibration methods have been proposed; however, most predict extrinsic parameters in a single step and lack iterative optimization capabilities. To address the increasing demand for higher accuracy, we propose a versatile iterative framework based on surrogate diffusion. This framework can enhance the performance of any calibration method without requiring architectural modifications. Specifically, the initial extrinsic parameters undergo iterative refinement through a denoising process, in which the original calibration method serves as a surrogate denoiser to estimate the final extrinsics at each step. For comparative analysis, we selected four state-of-the-art calibration methods as surrogate denoisers and compared the results of our diffusion process with those of two other iterative approaches. Extensive experiments demonstrate that when integrated with our diffusion model, all calibration methods achieve higher accuracy, improved robustness, and greater stability compared to other iterative techniques and their single-step counterparts. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:44:51 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Ou",
"Ni",
""
],
[
"Chen",
"Zhuo",
""
],
[
"Zhang",
"Xinru",
""
],
[
"Wang",
"Junzheng",
""
]
] |
2506.14707 | Qian Xu | Qian Xu, Feng Zhang, Chengxi Li, Lei Cao, Zheng Chen, Jidong Zhai and Xiaoyong Du | HARMONY: A Scalable Distributed Vector Database for High-Throughput Approximate Nearest Neighbor Search | null | null | null | null | cs.DB | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Approximate Nearest Neighbor Search (ANNS) is essential for various data-intensive applications, including recommendation systems, image retrieval, and machine learning. Scaling ANNS to handle billions of high-dimensional vectors on a single machine presents significant challenges in memory capacity and processing efficiency. To address these challenges, distributed vector databases leverage multiple nodes for the parallel storage and processing of vectors. However, existing solutions often suffer from load imbalance and high communication overhead, primarily due to traditional partition strategies that fail to effectively distribute the workload. In this paper, we introduce Harmony, a distributed ANNS system that employs a novel multi-granularity partition strategy, combining dimension-based and vector-based partition. This strategy ensures a balanced distribution of computational load across all nodes while effectively minimizing communication costs. Furthermore, Harmony incorporates an early-stop pruning mechanism that leverages the monotonicity of distance computations in dimension-based partition, resulting in significant reductions in both computational and communication overhead. We conducted extensive experiments on diverse real-world datasets, demonstrating that Harmony outperforms leading distributed vector databases, achieving 4.63 times throughput on average in four nodes and 58% performance improvement over traditional distribution for skewed workloads. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:45:42 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Xu",
"Qian",
""
],
[
"Zhang",
"Feng",
""
],
[
"Li",
"Chengxi",
""
],
[
"Cao",
"Lei",
""
],
[
"Chen",
"Zheng",
""
],
[
"Zhai",
"Jidong",
""
],
[
"Du",
"Xiaoyong",
""
]
] |
2506.14708 | Lukasz Stettner | Tomasz Rogala and {\L}ukasz Stettner | Discrete time shadow price revisited | null | null | null | null | math.OC | http://creativecommons.org/licenses/by/4.0/ | In the paper discrete time shadow price is constructed for the market with several assets with given bid and ask prices. Shadow price is the price such that the problem of optimal utility from terminal wealth on the market without transaction costs gives the same value function as in the case of bid and ask prices. In the paper we solve first static problem for two assets and then we construct the shadow price for dynamic model. Finally we present construction of the shadow price for several assets. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:47:57 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Rogala",
"Tomasz",
""
],
[
"Stettner",
"Łukasz",
""
]
] |
2506.14709 | Kunal Swami | Kunal Swami, Debtanu Gupta, Amrit Kumar Muduli, Chirag Jaiswal and Pankaj Kumar Bajpai | DiFuse-Net: RGB and Dual-Pixel Depth Estimation using Window Bi-directional Parallax Attention and Cross-modal Transfer Learning | Accepted in IROS 2025 | null | null | null | cs.CV cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Depth estimation is crucial for intelligent systems, enabling applications from autonomous navigation to augmented reality. While traditional stereo and active depth sensors have limitations in cost, power, and robustness, dual-pixel (DP) technology, ubiquitous in modern cameras, offers a compelling alternative. This paper introduces DiFuse-Net, a novel modality decoupled network design for disentangled RGB and DP based depth estimation. DiFuse-Net features a window bi-directional parallax attention mechanism (WBiPAM) specifically designed to capture the subtle DP disparity cues unique to smartphone cameras with small aperture. A separate encoder extracts contextual information from the RGB image, and these features are fused to enhance depth prediction. We also propose a Cross-modal Transfer Learning (CmTL) mechanism to utilize large-scale RGB-D datasets in the literature to cope with the limitations of obtaining large-scale RGB-DP-D dataset. Our evaluation and comparison of the proposed method demonstrates its superiority over the DP and stereo-based baseline methods. Additionally, we contribute a new, high-quality, real-world RGB-DP-D training dataset, named Dual-Camera Dual-Pixel (DCDP) dataset, created using our novel symmetric stereo camera hardware setup, stereo calibration and rectification protocol, and AI stereo disparity estimation method. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:49:27 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Swami",
"Kunal",
""
],
[
"Gupta",
"Debtanu",
""
],
[
"Muduli",
"Amrit Kumar",
""
],
[
"Jaiswal",
"Chirag",
""
],
[
"Bajpai",
"Pankaj Kumar",
""
]
] |
2506.14710 | Frank VanGessel | Francis G. VanGessel, Mitul Pandya | Deep Learning-Based Prediction of High Explosive Induced Fluid Dynamics | To be published in Physics of Fluids. 22 pages. 13 figures | null | 10.1063/5.0274946 | null | physics.flu-dyn | http://creativecommons.org/licenses/by/4.0/ | Underwater explosions produce complex fluid phenomena relevant to diverse applications including maritime engineering, medical therapeutics, and inertial confinement fusion. These systems exhibit multiphase flows, chemical kinetics, and highly compressible dynamics that challenge traditional computational approaches. Current hydrodynamic solvers, while accurate, are computationally expensive and non-differentiable, limiting their use in design optimization and real-time applications. Here we show that deep neural networks can predict underwater explosion-induced fluid dynamics 4,025 times faster than traditional solvers while maintaining mean absolute percent errors below 0.005\% across all fluid state variables. Our approach maps from explosive material thermodynamic parameters to the temporal evolution of shock fronts and material interfaces, enabling rapid prediction of system behavior for a broad range of ideal explosive materials. Feature importance analysis reveals that exponential decay parameters of the explosive equation of state are the primary drivers of system dynamics, uncovering a previously unknown relationship between thermodynamic compatibility and energy transfer efficiency at material interfaces. Furthermore, we demonstrate an inverse design framework that leverages the differentiability of our neural surrogate to perform parameter discovery, recovering unknown explosive material properties to within 1\% accuracy through gradient-based optimization. This combination of rapid inference, physical insight, and inverse design capabilities provides a route to engineering controlled fluid behavior in underwater explosive systems through material design. We anticipate our approach could enable new applications in defense systems, underwater manufacturing, and medical procedures where precise control of shock waves and bubble dynamics is essential. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:50:05 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"VanGessel",
"Francis G.",
""
],
[
"Pandya",
"Mitul",
""
]
] |
2506.14711 | Petar Bojovi\'c | Petar Bojovi\'c, Timon Hilker, Si Wang, Johannes Obermeyer, Marnix Barendregt, Dorothee Tell, Thomas Chalopin, Philipp M. Preiss, Immanuel Bloch, Titus Franz | High-fidelity collisional quantum gates with fermionic atoms | 16 pages, 18 figures | null | null | null | cond-mat.quant-gas cond-mat.str-el quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum simulations of electronic structure and strongly correlated quantum phases are widely regarded as among the most promising applications of quantum computing. These simulations require the accurate implementation of motion and entanglement of fermionic particles. Instead of the commonly applied costly mapping to qubits, fermionic quantum computers offer the prospect of directly implementing electronic structure problems. Ultracold neutral atoms have emerged as a powerful platform for spin-based quantum computing, but quantum information can also be processed via the motion of bosonic or fermionic atoms offering a distinct advantage by intrinsically conserving crucial symmetries like electron number. Here we demonstrate collisional entangling gates with fidelities up to 99.75(6)% and lifetimes of Bell states beyond $10\,s$ via the control of fermionic atoms in an optical superlattice. Using quantum gas microscopy, we characterize both spin-exchange and pair-tunneling gates locally, and realize a robust, composite pair-exchange gate, a key building block for quantum chemistry simulations. Our results enable the preparation of complex quantum states and advanced readout protocols for a new class of scalable analog-digital hybrid quantum simulators. Once combined with local addressing, they mark a key step towards a fully digital fermionic quantum computer based on the controlled motion and entanglement of fermionic neutral atoms. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:50:08 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Bojović",
"Petar",
""
],
[
"Hilker",
"Timon",
""
],
[
"Wang",
"Si",
""
],
[
"Obermeyer",
"Johannes",
""
],
[
"Barendregt",
"Marnix",
""
],
[
"Tell",
"Dorothee",
""
],
[
"Chalopin",
"Thomas",
""
],
[
"Preiss",
"Philipp M.",
""
],
[
"Bloch",
"Immanuel",
""
],
[
"Franz",
"Titus",
""
]
] |
2506.14712 | Thomas Brunner | Thomas Brunner, Jason D. Holt, David McKeen, Alexander Wright | Canadian Contributions to the Search for Neutrinoless Double Beta Decay | null | null | 10.1139/cjp-2024-0300 | null | nucl-ex | http://creativecommons.org/licenses/by/4.0/ | The search for neutrinoless double beta decay has internationally been recognized as the most promising approach to determine the Majorana nature of neutrinos. This hypothesized decay would, if observed, violate lepton number in weak interactions by two units, hence, prove the existence of physics beyond the Standard Model. Current experiments with sensitivity to neutrinoless double beta decay half lives of $10^{26}$ years did not observe such a decay and worldwide efforts are ongoing to deploy experiments with half-life sensitivities beyond $10^{28}$ years. Canadian groups have been involved in this search for more than four decades. This article summarizes the historical experimental efforts and describes current Canadian contributions to neutrinoless double beta decay searches and their theoretical interpretation. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:50:22 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Brunner",
"Thomas",
""
],
[
"Holt",
"Jason D.",
""
],
[
"McKeen",
"David",
""
],
[
"Wright",
"Alexander",
""
]
] |
2506.14713 | Victorien Desbois | Victorien Desbois and Ocan Sankur and Fran\c{c}ois Schwarzentruber | Linear Planar 3-SAT and Its Applications in Planning | null | null | null | null | cs.CC | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Several fragments of the satisfiability problem have been studied in the literature. Among these, Linear 3-SAT is a satisfaction problem in which each clause (viewed as a set of literals) intersects with at most one other clause; moreover, any pair of clauses have at most one literal in common. Planar 3-SAT is a fragment which requires that the so-called variable-clause graph is planar. Both fragments are NP-complete and have applications in encoding NP-hard planning problems. In this paper, we investigate the complexity and applications of the fragment obtained combining both features. We define Linear Planar 3-SAT and prove its NP-completeness. We also study the reconfiguration problem of Linear Planar 3-SAT and show that it is PSPACE-complete. As an application, we use these new results to prove the NP-completeness of Bounded Connected Multi-Agent Pathfinding and the PSPACE-completeness of Connected Multi-Agent Pathfinding in two-dimensional grids. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:51:03 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Desbois",
"Victorien",
""
],
[
"Sankur",
"Ocan",
""
],
[
"Schwarzentruber",
"François",
""
]
] |
2506.14714 | Egor Larionov | Egor Larionov, Igor Santesteban, Hsiao-yu Chen, Gene Lin, Philipp Herholz, Ryan Goldade, Ladislav Kavan, Doug Roble, Tuur Stuyck | SkinCells: Sparse Skinning using Voronoi Cells | null | null | null | null | cs.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For decades, efficient real-time skinning methods have played a crucial role in animating character rigs for visual effects and games. These methods remain a fundamental component of modern applications. However, animatable digital asset creation predominantly remains a manual process. Current automated tools often fall short of delivering the desired level of quality for intricate and complex geometries, requiring manual touch-ups. We propose a fully automatic and robust method for generating high quality skinning weights given a user-provided skeleton and mesh in A- or T-pose. Notably, our approach provides direct sparsity controls, limiting the number of bone influences per vertex, which is essential for efficient asset creation for large-scale mobile experiences with multiple concurrent users. Our method additionally addresses the need for level-of-detail (LoD) variations in performance-sensitive applications, which are exacerbated on mobile platforms. By optimizing weights in space rather than on discrete points, we enable a single optimization result to be seamlessly applied to all levels of detail of that asset or even variations of that asset. To achieve this, we introduce a novel parameterized family of functions called SkinCells. We demonstrate how our automatic method is able to robustly compute skinning weights in cases where biharmonic weight computation fails. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:51:36 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Larionov",
"Egor",
""
],
[
"Santesteban",
"Igor",
""
],
[
"Chen",
"Hsiao-yu",
""
],
[
"Lin",
"Gene",
""
],
[
"Herholz",
"Philipp",
""
],
[
"Goldade",
"Ryan",
""
],
[
"Kavan",
"Ladislav",
""
],
[
"Roble",
"Doug",
""
],
[
"Stuyck",
"Tuur",
""
]
] |
2506.14715 | Hamidah Oderinwale | Hamidah Oderinwale | Procedural Knowledge Libraries: Towards Executable (Research) Memory | 24 pages | null | null | null | cs.DL | http://creativecommons.org/licenses/by-sa/4.0/ | Procedural Knowledge Libraries (PKLs) are frameworks for capturing the full arc of scientific inquiry, not just its outcomes. Whereas traditional libraries store static end products, PKLs preserve the process that leads to those results, including hypotheses, failures, decisions, and iterations. By addressing the loss of tacit knowledge -- typically buried in notebooks, emails, or memory -- PKLs lay a foundation for reproducible, collaborative, and adaptive research. PKLs provide executable, version-controlled records that contextualize each step of a research process. For example, a researcher using Jupyter notebooks could share not just final outputs, but also the reasoning, discarded approaches, and intermediate analyses that informed them. This work proposes a framework for implementing PKLs within the Jupyter ecosystem, supported by a lens-based transformation model and procedural storage schema. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:52:12 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Oderinwale",
"Hamidah",
""
]
] |
2506.14716 | Yinghao Ye | Ying-Hao Ye, Jia-Qi Jiang, En-Ze Li, Wei Zhang, Da-Chuang Li, Zhi-Han Zhu, Dong-Sheng Ding, Bao-Sen Shi | All-optical convolution utilizing processing in memory based on a cold atomic ensemble | 6 pages, 3 figures, supplmental material appended | null | null | null | physics.app-ph physics.atom-ph physics.optics | http://creativecommons.org/licenses/by/4.0/ | Processing in memory (PIM) has received significant attention due to its high efficiency, low latency, and parallelism. In optical computation, coherent memory is a crucial infrastructure for PIM frameworks. This study presents an all-optical convolution experiment conducted within computational storage based on a cold atomic ensemble. By exploiting the light-atom phase transfer facilitated by the electromagnetically induced transparency, we demonstrated spiral phase contrast processing of photon images in memory, resulting in the edge enhancement of retrieved images recorded using time-correlated photon imaging. In particular, adopting state-of-the-art atomic techniques provides a coherent memory lifetime exceeding 320 us for PIM operations. Our results highlight the significant potential of cold atomic ensembles as computational storage for developing all-optical PIM systems. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:52:32 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Ye",
"Ying-Hao",
""
],
[
"Jiang",
"Jia-Qi",
""
],
[
"Li",
"En-Ze",
""
],
[
"Zhang",
"Wei",
""
],
[
"Li",
"Da-Chuang",
""
],
[
"Zhu",
"Zhi-Han",
""
],
[
"Ding",
"Dong-Sheng",
""
],
[
"Shi",
"Bao-Sen",
""
]
] |
2506.14717 | Aditya Sai Pranith Ayapilla | A. S. P. Ayapilla (Graduate School of Information Sciences, Tohoku University, Sendai, Japan), Y. Hattori (Institute of Fluid Science, Tohoku University, Sendai, Japan) and S. Le Diz\`es (Aix Marseille Universit\'e, CNRS, Centrale M\'editerran\'ee, IRPHE, Marseille, France) | Triangular instability of a strained Batchelor vortex | 39 pages, 27 figures | null | null | null | physics.flu-dyn | http://creativecommons.org/licenses/by/4.0/ | We investigate the triangular instability of a Batchelor vortex subjected to a stationary triangular strain field generated by three satellite vortices, in the presence of weak axial flow. The analysis combines theoretical predictions with numerical simulations. Theoretically, the instability arises from resonant coupling between two quasi-neutral Kelvin modes with azimuthal wavenumbers $m$ and $m+3$ with the background strain. Numerically, we solve the linearized Navier-Stokes equations around a quasi-steady base flow to identify the most unstable modes, and compare their growth rates and frequencies with theoretical predictions for a Reynolds number $Re = 10^4$ and a straining strength $\epsilon = 0.008$. In the absence of axial flow, only the mode pair $(m_A, m_B) = (-1,2)$ (and its symmetric counterpart) is unstable. However, we show that additional combinations such as $(0,3)$, $(1,4)$, and $(2,5)$, which are otherwise strongly damped by the critical layer in the absence of axial flow, also become unstable once axial flow exceeds a certain threshold, as the critical layer damping is significantly reduced. Furthermore, we show that the most unstable mode in the no-axial-flow case, originating from the second branch of $m = -1$ and the first branch of $m = 2$, becomes less unstable as axial flow increases. It is eventually overtaken by a mode from the first branches of both wavenumbers, which then remains the dominant unstable mode across a wide range of axial flow strengths and Reynolds numbers. A comprehensive instability diagram as a function of the axial flow parameter is presented. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:52:48 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Ayapilla",
"A. S. P.",
"",
"Graduate School of Information Sciences, Tohoku University, Sendai, Japan"
],
[
"Hattori",
"Y.",
"",
"Institute of Fluid Science, Tohoku University, Sendai, Japan"
],
[
"Dizès",
"S. Le",
"",
"Aix Marseille Université, CNRS, Centrale Méditerranée, IRPHE, Marseille, France"
]
] |
2506.14718 | Eric J. Korpela | Eric J. Korpela, David P. Anderson, Jeff Cobb, Matt Lebofsky, Wei Liu, and Dan Werthimer (Space Sciences Laboratory, University of California, Berkeley) | SETI@home: Data Acquisition and Front-End Processing | 21 pages, 7 figures, 5 tables. Accepted to AJ | null | null | null | astro-ph.IM cs.DC physics.data-an | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | SETI@home is a radio Search for Extraterrestrial Intelligence (SETI) project, looking for technosignatures in data recorded at multiple observatories from 1998 to 2020. Most radio SETI projects analyze data using dedicated processing hardware. SETI@home uses a different approach: time-domain data is distributed over the Internet to $\gt 10^{5}$ volunteered home computers, which analyze it. The large amount of computing power this affords ($\sim 10^{15}$ floating-point operations per second (FPOP/s)) allows us to increase the sensitivity and generality of our search in three ways. We use coherent integration, a technique in which data is transformed so that the power of drifting signals is confined to a single discrete Fourier transform (DFT) bin. We perform this coherent search over 123 000 Doppler drift rates in the range ($\pm$100 Hz s$^{-1}$). Second, we search for a variety of signal types, such as pulsed signals and arbitrary repeated waveforms. The analysis uses a range of DFT sizes, with frequency resolutions ranging from 0.075 Hz to 1221 Hz. The front end of SETI@home produces a set of detections that exceed thresholds in power and goodness of fit. We accumulated $\sim 1.2\times 10^{10}$ such detections. The back end of SETI@home takes these detections, identifies and removes radio frequency interference (RFI), and looks for groups of detections that are consistent with extraterrestrial origin and that persist over long timescales. This paper describes the front end of SETI@home and provides parameters for the primary data source, the Arecibo Observatory; the back end and its results are described in a companion paper. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:52:48 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Korpela",
"Eric J.",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Anderson",
"David P.",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Cobb",
"Jeff",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Lebofsky",
"Matt",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Liu",
"Wei",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Werthimer",
"Dan",
"",
"Space Sciences Laboratory, University of California, Berkeley"
]
] |
2506.14719 | Haley Duba-Sullivan | Haley Duba-Sullivan, Aniket Pramanik, Venkatakrishnan Singanallur, Amirkoushyar Ziabari | Plug-and-Play with 2.5D Artifact Reduction Prior for Fast and Accurate Industrial Computed Tomography Reconstruction | Submitted to Journal of Nondestructive Evaluation | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Cone-beam X-ray computed tomography (XCT) is an essential imaging technique for generating 3D reconstructions of internal structures, with applications ranging from medical to industrial imaging. Producing high-quality reconstructions typically requires many X-ray measurements; this process can be slow and expensive, especially for dense materials. Recent work incorporating artifact reduction priors within a plug-and-play (PnP) reconstruction framework has shown promising results in improving image quality from sparse-view XCT scans while enhancing the generalizability of deep learning-based solutions. However, this method uses a 2D convolutional neural network (CNN) for artifact reduction, which captures only slice-independent information from the 3D reconstruction, limiting performance. In this paper, we propose a PnP reconstruction method that uses a 2.5D artifact reduction CNN as the prior. This approach leverages inter-slice information from adjacent slices, capturing richer spatial context while remaining computationally efficient. We show that this 2.5D prior not only improves the quality of reconstructions but also enables the model to directly suppress commonly occurring XCT artifacts (such as beam hardening), eliminating the need for artifact correction pre-processing. Experiments on both experimental and synthetic cone-beam XCT data demonstrate that the proposed method better preserves fine structural details, such as pore size and shape, leading to more accurate defect detection compared to 2D priors. In particular, we demonstrate strong performance on experimental XCT data using a 2.5D artifact reduction prior trained entirely on simulated scans, highlighting the proposed method's ability to generalize across domains. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:52:57 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Duba-Sullivan",
"Haley",
""
],
[
"Pramanik",
"Aniket",
""
],
[
"Singanallur",
"Venkatakrishnan",
""
],
[
"Ziabari",
"Amirkoushyar",
""
]
] |
2506.14720 | Antonios Saravanos | Antonios Saravanos (1) ((1) New York University) | How Warm-Glow Alters the Usability of Technology | null | null | null | null | cs.HC cs.CY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As technology increasingly aligns with users' personal values, traditional models of usability, focused on functionality and specifically effectiveness, efficiency, and satisfaction, may not fully capture how people perceive and evaluate it. This study investigates how the warm-glow phenomenon, the positive feeling associated with doing good, shapes perceived usability. An experimental approach was taken in which participants evaluated a hypothetical technology under conditions designed to evoke either the intrinsic (i.e., personal fulfillment) or extrinsic (i.e., social recognition) dimensions of warm-glow. A Multivariate Analysis of Variance as well as subsequent follow-up analyses revealed that intrinsic warm-glow significantly enhances all dimensions of perceived usability, while extrinsic warm-glow selectively influences perceived effectiveness and satisfaction. These findings suggest that perceptions of usability extend beyond functionality and are shaped by how technology resonates with users' broader sense of purpose. We conclude by proposing that designers consider incorporating warm-glow into technology as a strategic design decision. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:54:06 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Saravanos",
"Antonios",
"",
"New York University"
]
] |
2506.14721 | Martin Bojowald | Martin Bojowald and Luis Martinez | Large effects from quantum reference frames | 14 pages, two figures | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Reference frames are used to parameterize measurements of physical effects, but since their practical realization uses material objects, they may affect observations performed in a combined quantum state of the measured system together with the frame. Here, a procedure is used that makes it possible to describe non-monotonic reference scales in a quantum treatment, revealing large quantum effects in the measured system whenever a reference frame encounters a turning point. Subtle quantum correlations in the combined state of system and frame, and more broadly the concept of relational quantum mechanics, can be tested via a characteristic and surprisingly large shift in the measured value. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:55:32 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Bojowald",
"Martin",
""
],
[
"Martinez",
"Luis",
""
]
] |
2506.14722 | Ankitha E. Bangera | Ankitha E Bangera ((1) Department of Electrical Engineering, Indian Institute of Technology Bombay, Powai, Mumbai, India) | A stochastic noise model based excess noise factor expressions for staircase avalanche photodiodes | This article derives the excess noise factor expressions for staircase APDs using a stochastic noise model (18 pages, 4 figures, preprint under submission) and can be used as a replacement to Capasso's expressions. Capasso's formula and its equivalents are incorrect, detailed in our article titled "Mathematical proof of errors in Capasso's excess noise factor formula... ." | null | null | null | eess.SP cs.IT math.IT math.PR physics.app-ph physics.comp-ph | http://creativecommons.org/licenses/by/4.0/ | Multistep staircase avalanche photodiodes (APDs) are the solid-state analogue of photomultiplier tubes, owing to their deterministic amplification with twofold stepwise gain via impact ionization. Yet, the stepwise impact ionization irregularities worsen with increasing step counts, which are a major source of internal noise in these APDs. Some noise models for staircase APDs have been previously reported, where the excess noise factor expressions are based on Friis' noise factor formula for cascade networks, erroneously considering the power gains as the gains. Excess noise factor being a key component in staircase APDs' noise models, we formulate generalized excess noise factor expressions for multilayer graded-bandgap APDs in terms of their layer-wise ionization probabilities, applicable for all operating biases, which include the sub-threshold, staircase, and tunnelling breakdown regimes. We further derive simplified expressions for staircase APDs and prove that these expressions match Bangera's corrections to Friis' noise factor formulas for cascade networks. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:58:44 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Bangera",
"Ankitha E",
""
]
] |
2506.14723 | Yusong Wu | Yusong Wu, Tim Cooijmans, Kyle Kastner, Adam Roberts, Ian Simon, Alexander Scarlatos, Chris Donahue, Cassie Tarakajian, Shayegan Omidshafiei, Aaron Courville, Pablo Samuel Castro, Natasha Jaques, Cheng-Zhi Anna Huang | Adaptive Accompaniment with ReaLchords | Accepted by ICML 2024 | null | null | null | cs.SD cs.AI | http://creativecommons.org/licenses/by/4.0/ | Jamming requires coordination, anticipation, and collaborative creativity between musicians. Current generative models of music produce expressive output but are not able to generate in an \emph{online} manner, meaning simultaneously with other musicians (human or otherwise). We propose ReaLchords, an online generative model for improvising chord accompaniment to user melody. We start with an online model pretrained by maximum likelihood, and use reinforcement learning to finetune the model for online use. The finetuning objective leverages both a novel reward model that provides feedback on both harmonic and temporal coherency between melody and chord, and a divergence term that implements a novel type of distillation from a teacher model that can see the future melody. Through quantitative experiments and listening tests, we demonstrate that the resulting model adapts well to unfamiliar input and produce fitting accompaniment. ReaLchords opens the door to live jamming, as well as simultaneous co-creation in other modalities. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:59:05 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Wu",
"Yusong",
""
],
[
"Cooijmans",
"Tim",
""
],
[
"Kastner",
"Kyle",
""
],
[
"Roberts",
"Adam",
""
],
[
"Simon",
"Ian",
""
],
[
"Scarlatos",
"Alexander",
""
],
[
"Donahue",
"Chris",
""
],
[
"Tarakajian",
"Cassie",
""
],
[
"Omidshafiei",
"Shayegan",
""
],
[
"Courville",
"Aaron",
""
],
[
"Castro",
"Pablo Samuel",
""
],
[
"Jaques",
"Natasha",
""
],
[
"Huang",
"Cheng-Zhi Anna",
""
]
] |
2506.14724 | Faisal Etminan | Faisal Etminan | Quantitative predictions of alpha-charmonium correlation functions in high-energy collisions | 16 pages, 4 figures, 2 Tables | null | null | RIKEN-iTHEMS-Report-25 | nucl-th | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Two-body $ ^{4}{He}\left(\alpha\right)$-charmonium $ \left(c\bar{c}\right) $ potentials in the single-folding potential (SFP) approach are built by using a first principles HAL QCD low-energy $ NJ/\psi$ and $ N\eta_{c} $ interactions. It is found that the $ \alpha-c\bar{c} $ system based on the spin-$ 1/2 $ $NJ/\psi $ appears to be loosely bound with a binding energy of $ \sim 0.1 $ MeV, while for those based on spin-$3/2$ $ NJ/\psi $, spin-$ 1/2 $ $N\eta_{c}$ and the spin-averaged $ NJ/\psi $, no bound or resonance state (with respect to the $ \alpha + c\bar{c} $ threshold) was found. The $ \alpha c\bar{c} $ correlation function in high-energy collisions is examined to explore the $ N c\bar{c} $ interaction. The analysis revealed that variations in spin-dependent $N c\bar{c}$ interactions, spin-$3/2$ $NJ/\psi$, spin-$1/2$ $NJ/\psi$, spin-$1/2$ $N\eta_c$, and the spin-averaged $NJ/\psi$,produce noticeable differences in the $\alpha c\bar{c}$ correlation function, especially when the source size is around $ 3 $ fm. It is found that different results are produced by the Lednicky-Lyuboshits formula at small source sizes. This indicates that a relatively long-range interaction exists for the $ \alpha c\bar{c} $ system. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 16:59:12 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Etminan",
"Faisal",
""
]
] |
2506.14725 | Mark Huber | Mark Huber | Generating uniform linear extensions using few random bits | 16 pages | null | null | null | cs.CC math.PR stat.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A \emph{linear extension} of a partial order \(\preceq\) over items \(A = \{ 1, 2, \ldots, n \}\) is a permutation \(\sigma\) such that for all \(i < j\) in \(A\), it holds that \(\neg(\sigma(j) \preceq \sigma(i))\). Consider the problem of generating uniformly from the set of linear extensions of a partial order. The best method currently known uses \(O(n^3 \ln(n))\) operations and \(O(n^3 \ln(n)^2)\) iid fair random bits to generate such a permutation. This paper presents a method that generates a uniform linear extension using only \(2.75 n^3 \ln(n)\) operations and \( 1.83 n^3 \ln(n) \) iid fair bits on average. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:02:42 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Huber",
"Mark",
""
]
] |
2506.14726 | Elena Savchenko | M.A. Bludov, I.V. Khyzhniy, S.A. Uyutnov, E.V. Savchenko | About explosive delayed desorption from methane-doped argon matrices | 16 pages, 10 figures | Low Temp. Phys., v.51 (2025) 489-496 | 10.1063/10.0036208 | null | physics.chem-ph | http://creativecommons.org/licenses/by/4.0/ | The features of delayed desorption from CH4-doped Ar matrices irradiated with an electron beam of subthreshold energy were studied. Radiolysis products were detected by emission spectroscopy. The total desorption yield was monitored by recording the pressure in the experimental chamber. Based on the analysis of the concentration dependence of delayed desorption bursts and their structure, an assumption was made about the formation of CH4 clusters in Ar matrices. At a high dopant concentration of 10%, up to three consecutive bursts were recorded. Delayed desorption from a sample doped with 0.1% CH4 was registered for the first time. A correlation has been found between the burst of particles and the flash of luminescence of H atoms. This seemingly contradictory observation of the correlation of the H atom content with the particle explosion due to their recombination was explained by analyzing the energy transfer and capture processes and the features of the energy structure of H2 molecule. A linear dependence of the total particle yield on the CH4 concentration in the range of 1-10% was found. The dynamics of intensity changes in the sequence of main flashes and their delay time were discussed. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:02:56 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Bludov",
"M. A.",
""
],
[
"Khyzhniy",
"I. V.",
""
],
[
"Uyutnov",
"S. A.",
""
],
[
"Savchenko",
"E. V.",
""
]
] |
2506.14727 | Huihan Liu | Huihan Liu, Rutav Shah, Shuijing Liu, Jack Pittenger, Mingyo Seo, Yuchen Cui, Yonatan Bisk, Roberto Mart\'in-Mart\'in, Yuke Zhu | Casper: Inferring Diverse Intents for Assistive Teleoperation with Vision Language Models | null | null | null | null | cs.RO cs.AI | http://creativecommons.org/licenses/by/4.0/ | Assistive teleoperation, where control is shared between a human and a robot, enables efficient and intuitive human-robot collaboration in diverse and unstructured environments. A central challenge in real-world assistive teleoperation is for the robot to infer a wide range of human intentions from user control inputs and to assist users with correct actions. Existing methods are either confined to simple, predefined scenarios or restricted to task-specific data distributions at training, limiting their support for real-world assistance. We introduce Casper, an assistive teleoperation system that leverages commonsense knowledge embedded in pre-trained visual language models (VLMs) for real-time intent inference and flexible skill execution. Casper incorporates an open-world perception module for a generalized understanding of novel objects and scenes, a VLM-powered intent inference mechanism that leverages commonsense reasoning to interpret snippets of teleoperated user input, and a skill library that expands the scope of prior assistive teleoperation systems to support diverse, long-horizon mobile manipulation tasks. Extensive empirical evaluation, including human studies and system ablations, demonstrates that Casper improves task performance, reduces human cognitive load, and achieves higher user satisfaction than direct teleoperation and assistive teleoperation baselines. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:06:43 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Liu",
"Huihan",
""
],
[
"Shah",
"Rutav",
""
],
[
"Liu",
"Shuijing",
""
],
[
"Pittenger",
"Jack",
""
],
[
"Seo",
"Mingyo",
""
],
[
"Cui",
"Yuchen",
""
],
[
"Bisk",
"Yonatan",
""
],
[
"Martín-Martín",
"Roberto",
""
],
[
"Zhu",
"Yuke",
""
]
] |
2506.14728 | Xinzhe Juan | Jiahao Qiu, Xinzhe Juan, Yimin Wang, Ling Yang, Xuan Qi, Tongcheng Zhang, Jiacheng Guo, Yifu Lu, Zixin Yao, Hongru Wang, Shilong Liu, Xun Jiang, Liu Leqi, and Mengdi Wang | AgentDistill: Training-Free Agent Distillation with Generalizable MCP Boxes | 10 pages, 5 figures | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | While knowledge distillation has become a mature field for compressing large language models (LLMs) into smaller ones by aligning their outputs or internal representations, the distillation of LLM-based agents, which involve planning, memory, and tool use, remains relatively underexplored. Existing agent distillation methods typically replay full teacher trajectories or imitate step-by-step teacher tool usage, but they often struggle to train student agents to dynamically plan and act in novel environments. We propose AgentDistill, a novel, training-free agent distillation framework that enables efficient and scalable knowledge transfer via direct reuse of Model-Context-Protocols (MCPs), which are structured and reusable task-solving modules autonomously generated by teacher agents. The reuse of these distilled MCPs enables student agents to generalize their capabilities across domains and solve new problems with minimal supervision or human intervention. Experiments on biomedical and mathematical benchmarks demonstrate that our distilled student agents, built on small language models, can achieve performance comparable to advanced systems using large LLMs such as OctoTools (GPT-4o), highlighting the effectiveness of our framework in building scalable and cost-efficient intelligent agents. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:08:32 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Qiu",
"Jiahao",
""
],
[
"Juan",
"Xinzhe",
""
],
[
"Wang",
"Yimin",
""
],
[
"Yang",
"Ling",
""
],
[
"Qi",
"Xuan",
""
],
[
"Zhang",
"Tongcheng",
""
],
[
"Guo",
"Jiacheng",
""
],
[
"Lu",
"Yifu",
""
],
[
"Yao",
"Zixin",
""
],
[
"Wang",
"Hongru",
""
],
[
"Liu",
"Shilong",
""
],
[
"Jiang",
"Xun",
""
],
[
"Leqi",
"Liu",
""
],
[
"Wang",
"Mengdi",
""
]
] |
2506.14729 | Maurizio Quadrio | Marco Atzori, Emanuele Gallorini, Ciro Cottini, Andrea Benassi and Maurizio Quadrio | Direct numerical simulations of inhalation in a 23-generation lung model | null | null | null | null | physics.flu-dyn | http://creativecommons.org/licenses/by-nc-nd/4.0/ | The air flows in the proximal and distal portions of the human lungs are interconnected: the lower Reynolds number in the deeper generations causes a progressive flow regularization, while mass conservation requires flow rate oscillations to propagate through the airway bifurcations. To explain how these two competing effects shape the flow state in the deeper generations, we have performed the first high-fidelity numerical simulations of the air flow in a lung model including 23 successive bifurcations of a single planar airway. Turbulence modelling or assumptions on flow regimes are not required. The chosen flow rate is stationary (steady on average), and representative of the peak inspiratory flow reached by adult patients breathing through therapeutical inhalers. As expected, advection becomes progressively less important after each bifurcation, until a time-dependent Stokes regime governed solely by viscous diffusion is established in the smallest generations. However, fluctuations in this regime are relatively fast and large with respect to the mean flow, which is in contrast with the commonly agreed picture that only the breathing frequency is relevant at the scale of the alveoli. We demonstrate that the characteristic frequency and amplitude of these fluctuations are linked to the flow in the upper part of the bronchial tree, as they originate from the time-dependent flow splitting in the upper bifurcations. Even though these fluctuations are observed here in an idealized, rigid lung model, our findings suggest that the assumptions usually adopted in many of the current lung models might need to be revised. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:12:08 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Atzori",
"Marco",
""
],
[
"Gallorini",
"Emanuele",
""
],
[
"Cottini",
"Ciro",
""
],
[
"Benassi",
"Andrea",
""
],
[
"Quadrio",
"Maurizio",
""
]
] |
2506.14730 | Corey Scher | Corey Scher and Jamon Van Den Hoek | Active InSAR monitoring of building damage in Gaza during the Israel-Hamas War | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Aerial bombardment of the Gaza Strip beginning October 7, 2023 is one of the most intense bombing campaigns of the twenty-first century, driving widespread urban damage. Characterizing damage over a geographically dynamic and protracted armed conflict requires active monitoring. Synthetic aperture radar (SAR) has precedence for mapping disaster-induced damage with bi-temporal methods but applications to active monitoring during sustained crises are limited. Using interferometric SAR data from Sentinel-1, we apply a long temporal-arc coherent change detection (LT-CCD) approach to track weekly damage trends over the first year of the 2023- Israel-Hamas War. We detect 92.5% of damage labels in reference data from the United Nations with a negligible (1.2%) false positive rate. The temporal fidelity of our approach reveals rapidly increasing damage during the first three months of the war focused in northern Gaza, a notable pause in damage during a temporary ceasefire, and surges of new damage as conflict hot-spots shift from north to south. Three-fifths (191,263) of all buildings are damaged or destroyed by the end of the study. With massive need for timely data on damage in armed conflict zones, our low-cost and low-latency approach enables rapid uptake of damage information at humanitarian and journalistic organizations. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:12:22 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Scher",
"Corey",
""
],
[
"Hoek",
"Jamon Van Den",
""
]
] |
2506.14732 | Stefan Schr\"oer | Stefan Schr\"oer | K3 surfaces over small number fields and Kummer constructions in families | 40 pages | null | null | null | math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We construct K3 surfaces over number fields that have good reduction everywhere. These do not exists over the rational numbers, by results of Abrashkin and Fontaine. Our surfaces exist for three quadratic number fields, and an infinite family of $S_3$-number fields. To this end we develop a theory of Kummer constructions in families, based on Romagny's notion of the effective models, here applied to sign involutions. This includes quotients of non-normal surfaces by infinitesimal group schemes in characteristic two, as developed by Kondo and myself. By the results of Brieskorn and Artin, the resulting families of normal K3 surfaces admit simultaneous resolutions of singularities, at least after suitable base-changes. These resolutions are constructed in two ways: First, by blowing-up families of one-dimensional centers that acquire embedded components. Second, by computing various l-adic local systems in terms of representation theory, and invoking Shepherd-Barrons results on the resolution functor, which is representable by a highly non-separated algebraic space. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:14:29 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Schröer",
"Stefan",
""
]
] |
2506.14733 | Kamal Kumar Paul PhD | Kamal Kumar Paul (1), Cullen Chosy (2 and 3), Soumya Sarkar (1), Zhuangnan Li (1), Han Yan (1), Ye Wang (1), Leyi Loh (1), Lixin Liu (1), Hu Young Jeong (4), Samuel D. Stranks (2 and 3), Yan Wang (1), and Manish Chhowalla (1) | High-efficiency WSe$_2$ photovoltaics enabled by ultra-clean van der Waals contacts | Main manuscript with four figures and supplementary figures | null | null | null | cond-mat.mtrl-sci physics.app-ph | http://creativecommons.org/licenses/by/4.0/ | Layered transition metal dichalcogenide semiconductors are interesting for photovoltaics owing to their high solar absorbance and efficient carrier diffusion. Tungsten diselenide (WSe$_2$), in particular, has emerged as a promising solar cell absorber. However, defective metal-semiconductor interfaces have restricted the power conversion efficiency (PCE) to approximately 6%. Here we report WSe$_2$ photovoltaics with a record-high PCE of approximately 11% enabled by ultra-clean indium/gold (In/Au) van der Waals (vdW) contacts. Using grid-patterned top vdW electrodes, we demonstrate near-ideal diodes with a record-high on/off ratio of $1.0\times 10^9$. Open-circuit voltage (VOC) of 571 +/- 9 mV, record-high short-circuit current density (JSC) of 27.19 +/- 0.45 mA cm$^{-2}$ -- approaching the theoretical limit (34.5 mA cm$^{-2}$) -- and fill factor of 69.2 +/- 0.7% resulting in PCE of 10.8 +/- 0.2% under 1-Sun illumination on large active area (approximately 0.13x0.13 mm$^2$) devices have been realised. The excellent device performance is consistent with the high external quantum efficiency (up to approximately 93%) measured across a broad spectral range of 500-830 nm. Our results suggest that ultra-clean vdW contacts on WSe$_2$ enable high-efficiency photovoltaics and form the foundation for further optimisation. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:14:48 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Paul",
"Kamal Kumar",
"",
"2 and 3"
],
[
"Chosy",
"Cullen",
"",
"2 and 3"
],
[
"Sarkar",
"Soumya",
"",
"2 and 3"
],
[
"Li",
"Zhuangnan",
"",
"2 and 3"
],
[
"Yan",
"Han",
"",
"2 and 3"
],
[
"Wang",
"Ye",
"",
"2 and 3"
],
[
"Loh",
"Leyi",
"",
"2 and 3"
],
[
"Liu",
"Lixin",
"",
"2 and 3"
],
[
"Jeong",
"Hu Young",
"",
"2 and 3"
],
[
"Stranks",
"Samuel D.",
"",
"2 and 3"
],
[
"Wang",
"Yan",
""
],
[
"Chhowalla",
"Manish",
""
]
] |
2506.14734 | Nicola Prezza | Ruben Becker, Davide Cenzato, Travis Gagie, Sung-Hwan Kim, Ragnar Groot Koerkamp, Giovanni Manzini, Nicola Prezza | Compressing Suffix Trees by Path Decompositions | preliminary incomplete draft. Many details missing! | null | null | null | cs.DS | http://creativecommons.org/licenses/by/4.0/ | In classic suffix trees, path compression works by replacing unary suffix trie paths with pairs of pointers to $T$, which must be available in the form of some random access oracle at query time. In this paper, we revisit path compression and show that a more careful choice of pointers leads to a new elegant, simple, and remarkably efficient way to compress the suffix tree. We begin by observing that an alternative way to path-compress the suffix trie of $T$ is to decompose it into a set of (disjoint) node-to-leaf paths and then represent each path as a pointer $i$ to one of the string's suffixes $T[i,n]$. At this point, we show that the array $A$ of such indices $i$, sorted by the colexicographic order of the corresponding text prefixes $T[1,i]$, possesses the following properties: (i) it supports \emph{cache-efficient} pattern matching queries via simple binary search on $A$ and random access on $T$, and (ii) it contains a number of entries being proportional to the size of the \emph{compressed text}. Of particular interest is the path decomposition given by the colexicographic rank of $T$'s prefixes. The resulting index is smaller and orders of magnitude faster than the $r$-index on the task of locating all occurrences of a query pattern. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:15:08 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Becker",
"Ruben",
""
],
[
"Cenzato",
"Davide",
""
],
[
"Gagie",
"Travis",
""
],
[
"Kim",
"Sung-Hwan",
""
],
[
"Koerkamp",
"Ragnar Groot",
""
],
[
"Manzini",
"Giovanni",
""
],
[
"Prezza",
"Nicola",
""
]
] |
2506.14735 | Deping Ye | Xiao Li, Nguyen Nguyen and Deping Ye | A Minkowski problem for $\alpha$-concave functions via optimal transport | null | null | null | null | math.FA math.AP math.MG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The notions of the Euclidean surface area measure and the spherical surface area measure of $\alpha$-concave functions in $\mathbb{R}^n$, with $-\frac{1}{n}<\alpha<0$, are introduced via a first variation of the total mass functional with respect to the $\alpha$-sum operation. Subsequently, these notions are extended to those for $\alpha$-concave measures. We then study the Minkowski problem associated with the Euclidean surface area measures of $\alpha$-concave measures via optimal transport. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:17:48 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Li",
"Xiao",
""
],
[
"Nguyen",
"Nguyen",
""
],
[
"Ye",
"Deping",
""
]
] |
2506.14736 | Enrique P\'erez-Montero Dr | Enrique P\'erez-Montero, Borja P\'erez-D\'iaz, Jos\'e M. V\'ilchez, Igor A. Zinchenko, Asier Castrillo, Marta Gavil\'an, Sandra Zamora, and \'Angeles I. D\'iaz | Sulphur abundances in star-forming regions from optical emission lines: A new approach based on photoionization models consistent with the direct method | 13 pages, 9 figures, to be submitted to The Open Journal of Astrophysics | null | null | null | astro-ph.GA | http://creativecommons.org/licenses/by/4.0/ | The derivation of sulphur chemical abundances in the gas-phase of star-forming galaxies is explored in this work, using the emission lines produced in these regions in the optical part of the spectrum and by means of photoionization models. We adapted the code HII-CHI-mistry to account for these abundances by implementing additional grids of models that assume a variable sulphur-to-oxygen abundance ratio, beyond the commonly assumed solar ratio. The addition of these models, and their use in a new iteration of the code allows us to use sulphur lines to precisely estimate the sulphur abundance, even in the absence of auroral lines. This approach aligns with the results from the direct method, and no additional assumptions about the ionization correction factor are needed, as the models directly predict the total sulphur abundance. We applied this new methodology to a large sample of star-forming regions from the MaNGA survey, and we explored the variation of the S/O ratio as a function of metallicity, making corrections for the significant contribution from diffuse ionized gas, which particularly affects the [SII] emission. Our results indicate no significant deviations from the solar S/O value in the range 8.0 < 12+log(O/H) < 8.7, where the bulk of the MaNGA sample stays, , but also with possible enhancements of sulphur production both at low and high metallicity regimes. The latter may be linked to the depletion of oxygen in the gas-phase due to its incorporation onto dust grains | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:18:20 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Pérez-Montero",
"Enrique",
""
],
[
"Pérez-Díaz",
"Borja",
""
],
[
"Vílchez",
"José M.",
""
],
[
"Zinchenko",
"Igor A.",
""
],
[
"Castrillo",
"Asier",
""
],
[
"Gavilán",
"Marta",
""
],
[
"Zamora",
"Sandra",
""
],
[
"Díaz",
"Ángeles I.",
""
]
] |
2506.14737 | Eric J. Korpela | David P. Anderson (1), Eric J. Korpela (1), Dan Werthimer (1), Jeff Cobb (1) and Bruce Allen (2) ((1) Space Sciences Laboratory, University of California, Berkeley, (2) Max Planck Institut f\"ur Gravitationsphysik (Albert Einstein Institut), Hanover) | SETI@home: Data Analysis and Findings | 46 pages, 9 appendices, 20 figures, Accepted to AJ | null | null | null | astro-ph.IM physics.data-an | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | SETI@home is a radio Search for Extraterrestrial Intelligence (SETI) project that looks for technosignatures in data recorded at the Arecibo Observatory. The data were collected over a period of 14 years and cover almost the entire sky visible to the telescope. The first stage of data analysis found billions of detections: brief excesses of continuous or pulsed narrowband power. The second stage removed detections that were likely radio frequency interference (RFI), then identified and ranked signal candidates: groups of detections, possibly spread over the 14 years, that plausibly originate from a single cosmic source. We manually examined the top-ranking signal candidates and selected a few hundred. In the third and final stage we are reobserving the corresponding sky locations and frequency ranges using the Five-hundred-meter Aperture Spherical Telescope (FAST) radio telescope. This paper covers SETI@home's second stage of data analysis. We describe the algorithms used to remove RFI and to identify and rank signal candidates. To guide the development of these algorithms, we used artificial candidate birdies that model persistent ET signals with a range of power, bandwidth, and planetary motion parameters. This approach also allowed us to estimate the sensitivity of our detection system to these signals. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:18:55 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Anderson",
"David P.",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Korpela",
"Eric J.",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Werthimer",
"Dan",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Cobb",
"Jeff",
"",
"Space Sciences Laboratory, University of California, Berkeley"
],
[
"Allen",
"Bruce",
"",
"Max Planck Institut für Gravitationsphysik"
]
] |
2506.14738 | Matthias Allard | Matthias Allard, Peter J. Forrester, Sampad Lahiry, Bojian Shen | Partition function of 2D Coulomb gases with radially symmetric potentials and a hard wall | 30 pages | null | null | null | math.PR math-ph math.MP | http://creativecommons.org/licenses/by/4.0/ | The large $N$ asymptotic expansion of the partition function for the normal matrix model is predicted to have special features inherited from its interpretation as a two-dimensional Coulomb gas. However for the latter, it is most natural to include a hard wall at the boundary of the droplet. We probe how this affects the asymptotic expansion in the solvable case that the potential is radially symmetric and the droplet is a disk or an annulus. We allow too for the hard wall to be strictly inside the boundary of the droplet. It is observed the term of order $\log N$, has then a different rational number prefactor to that when the hard wall is at the droplet boundary. Also found are certain universal (potential independent) numerical constants given by definite integrals, both at order $\sqrt{N}$, and in the constant term. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:19:33 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Allard",
"Matthias",
""
],
[
"Forrester",
"Peter J.",
""
],
[
"Lahiry",
"Sampad",
""
],
[
"Shen",
"Bojian",
""
]
] |
2506.14739 | Jenna Elliott | Jenna Elliott, Hiral Shah, Roman Belousov, Gautam Dey and Anna Erzberger | Repulsive particle interactions enable selective information processing at cellular interfaces | 23 pages, 6 figures | null | null | null | cond-mat.stat-mech physics.bio-ph | http://creativecommons.org/licenses/by/4.0/ | Living systems relay information across membrane interfaces to coordinate compartment functions. We identify a physical mechanism for selective information transmission that arises from the sigmoidal response of surface-bound particle densities to spatial features of adjacent external structures. This mechanism implements a form of spatial thresholding, enabling the binary classification of external cues. Expansion microscopy measurements of nuclear pore complex distributions in S. arctica show signatures of such physical thresholding. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:20:01 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Elliott",
"Jenna",
""
],
[
"Shah",
"Hiral",
""
],
[
"Belousov",
"Roman",
""
],
[
"Dey",
"Gautam",
""
],
[
"Erzberger",
"Anna",
""
]
] |
2506.14740 | Chin-Wen Chou | Dalton Chaffee, Baruch Margulis, April Sheffield, Julian Schmidt, April Reisenfeld, David R. Leibrandt, Dietrich Leibfried, Chin-Wen Chou | High-fidelity quantum state control of a polar molecular ion in a cryogenic environment | null | null | null | null | physics.atom-ph quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We use a quantum-logic spectroscopy (QLS) protocol to control the quantum state of a CaH+ ion in a cryogenic environment, in which reduced thermal radiation extends rotational state lifetimes by an order of magnitude over those at room temperature. By repeatedly and adaptively probing the molecule, detecting the outcome of each probe via an atomic ion, and using a Bayesian update scheme to quantify confidence in the molecular state, we demonstrate state preparation and measurement (SPAM) in a single quantum state with infidelity less than $6\times10^{-3}$ and measure Rabi flopping between two states with greater than 99% contrast. The protocol does not require any molecule-specific lasers and the detection scheme is non-destructive. To our knowledge, this result represents the highest SPAM fidelity of a single molecular quantum state demonstrated to date. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:20:10 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Chaffee",
"Dalton",
""
],
[
"Margulis",
"Baruch",
""
],
[
"Sheffield",
"April",
""
],
[
"Schmidt",
"Julian",
""
],
[
"Reisenfeld",
"April",
""
],
[
"Leibrandt",
"David R.",
""
],
[
"Leibfried",
"Dietrich",
""
],
[
"Chou",
"Chin-Wen",
""
]
] |
2506.14741 | Ian Petrow | Yueke Hu and Ian Petrow and Matthew P. Young | The cubic moment of $L$-functions for specified local component families | null | null | null | null | math.NT | http://creativecommons.org/licenses/by/4.0/ | We prove Lindel\"of-on-average upper bounds on the cubic moment of central values of $L$-functions over certain families of ${\rm PGL}_2/\mathbb{Q}$ automorphic representations $\pi$ given by specifying the local representation $\pi_p$ of $\pi$ at finitely many primes. Such bounds were previously known in the case that $\pi_p$ belongs to the principal series or is a ramified quadratic twist of the Steinberg representation; here we handle the supercuspidal case. Crucially, we use new Petersson/Bruggeman/Kuznetsov forumulas for supercuspidal local component families recently developed by the authors. As corollaries, we derive Weyl-strength subconvex bounds for central values of ${\rm PGL}_2$ $L$-functions in the square-full aspect, and in the depth aspect, or in a hybrid of these two situations. A special case of our results is the Weyl-subconvex bound for all cusp forms of level $p^2$. Previously, such a bound was only known for forms that are twists from level $p$, which cover roughly half of the level $p^2$ forms. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:21:00 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Hu",
"Yueke",
""
],
[
"Petrow",
"Ian",
""
],
[
"Young",
"Matthew P.",
""
]
] |
2506.14742 | Ziqiao Peng | Ziqiao Peng, Wentao Hu, Junyuan Ma, Xiangyu Zhu, Xiaomei Zhang, Hao Zhao, Hui Tian, Jun He, Hongyan Liu, Zhaoxin Fan | SyncTalk++: High-Fidelity and Efficient Synchronized Talking Heads Synthesis Using Gaussian Splatting | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Achieving high synchronization in the synthesis of realistic, speech-driven talking head videos presents a significant challenge. A lifelike talking head requires synchronized coordination of subject identity, lip movements, facial expressions, and head poses. The absence of these synchronizations is a fundamental flaw, leading to unrealistic results. To address the critical issue of synchronization, identified as the ''devil'' in creating realistic talking heads, we introduce SyncTalk++, which features a Dynamic Portrait Renderer with Gaussian Splatting to ensure consistent subject identity preservation and a Face-Sync Controller that aligns lip movements with speech while innovatively using a 3D facial blendshape model to reconstruct accurate facial expressions. To ensure natural head movements, we propose a Head-Sync Stabilizer, which optimizes head poses for greater stability. Additionally, SyncTalk++ enhances robustness to out-of-distribution (OOD) audio by incorporating an Expression Generator and a Torso Restorer, which generate speech-matched facial expressions and seamless torso regions. Our approach maintains consistency and continuity in visual details across frames and significantly improves rendering speed and quality, achieving up to 101 frames per second. Extensive experiments and user studies demonstrate that SyncTalk++ outperforms state-of-the-art methods in synchronization and realism. We recommend watching the supplementary video: https://ziqiaopeng.github.io/synctalk++. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:22:12 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Peng",
"Ziqiao",
""
],
[
"Hu",
"Wentao",
""
],
[
"Ma",
"Junyuan",
""
],
[
"Zhu",
"Xiangyu",
""
],
[
"Zhang",
"Xiaomei",
""
],
[
"Zhao",
"Hao",
""
],
[
"Tian",
"Hui",
""
],
[
"He",
"Jun",
""
],
[
"Liu",
"Hongyan",
""
],
[
"Fan",
"Zhaoxin",
""
]
] |
2506.14743 | Sergio Iserte | Sergio Iserte, Iker Mart\'in-\'Alvarez, Krzysztof Rojek, Jos\'e I. Aliaga, Maribel Castillo, Weronika Folwarska, and Antonio J. Pe\~na | Resource Optimization with MPI Process Malleability for Dynamic Workloads in HPC Clusters | null | Future Generation Computer Systems, p. 107949, 2025 | 10.1016/j.future.2025.107949 | null | cs.DC | http://creativecommons.org/licenses/by/4.0/ | Dynamic resource management is essential for optimizing computational efficiency in modern high-performance computing (HPC) environments, particularly as systems scale. While research has demonstrated the benefits of malleability in resource management systems (RMS), the adoption of such techniques in production environments remains limited due to challenges in standardization, interoperability, and usability. Addressing these gaps, this paper extends our prior work on the Dynamic Management of Resources (DMR) framework, which provides a modular and user-friendly approach to dynamic resource allocation. Building upon the original DMRlib reconfiguration runtime, this work integrates new methodology from the Malleability Module (MaM) of the Proteo framework, further enhancing reconfiguration capabilities with new spawning strategies and data redistribution methods. In this paper, we explore new malleability strategies in HPC dynamic workloads, such as merging MPI communicators and asynchronous reconfigurations, which offer new opportunities for dramatically reducing memory overhead. The proposed enhancements are rigorously evaluated on a world-class supercomputer, demonstrating improved resource utilization and workload efficiency. Results show that dynamic resource management can reduce the workload completion time by 40% and increase the resource utilization by over 20%, compared to static resource allocation. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:28:31 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Iserte",
"Sergio",
""
],
[
"Martín-Álvarez",
"Iker",
""
],
[
"Rojek",
"Krzysztof",
""
],
[
"Aliaga",
"José I.",
""
],
[
"Castillo",
"Maribel",
""
],
[
"Folwarska",
"Weronika",
""
],
[
"Peña",
"Antonio J.",
""
]
] |
2506.14744 | Eleanor Gallay | Eleanor M. Gallay, James R. A. Davenport, and Steve Croft | Technosignature Searches with Real-time Alert Brokers | null | null | 10.3847/1538-3881/ade4bb | null | astro-ph.IM astro-ph.SR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present an exploration of technosignature research that is possible using real-time alert brokers from surveys such as the Zwicky Transient Facility (ZTF) and the upcoming Legacy Survey of Space and Time (LSST). Nine alert brokers currently stream up to 1 million alerts each night from ZTF, and LSST is projected to increase this volume by an order of magnitude. While these brokers are primarily designed to facilitate real-time follow-up of explosive transients such as supernovae, they offer a unique platform to discover rare forms of variability from nearby stars in real time, which is crucial for follow-up and characterization. We evaluate the capability for both spatial and temporal searches for extraterrestrial intelligence (SETI) methods using the currently available brokers, and present examples of technosignature searches using ZTF alert and archival data. We have deployed optical SETI techniques, such as planetary transit zone geometries and the SETI Ellipsoid. We have also developed a search for novel high-amplitude stellar dippers, and present a workflow that integrates features available directly through the brokers, as well as post-processing steps that build on the existing capabilities. Though the SETI methods that alert brokers can execute are currently limited, we provide suggestions that may enhance future technosignature and anomaly searches in the era of the Vera C. Rubin Observatory. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:29:23 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Gallay",
"Eleanor M.",
""
],
[
"Davenport",
"James R. A.",
""
],
[
"Croft",
"Steve",
""
]
] |
2506.14745 | Ching-Yi Lai | Tzu-Hao Lin and Ching-Yi Lai | Union-Intersection Union-Find for Decoding Depolarizing Errors in Topological Codes | 13 pages, 17 figures | null | null | null | quant-ph cs.IT math.IT | http://creativecommons.org/licenses/by/4.0/ | In this paper, we introduce the Union-Intersection Union-Find (UIUF) algorithm for decoding depolarizing errors in topological codes, combining the strengths of iterative and standard Union-Find (UF) decoding. While iterative UF improves performance at moderate error rates, it lacks an error correction guarantee. To address this, we develop UIUF, which maintains the enhanced performance of iterative UF while ensuring error correction up to half the code distance. Through simulations under code capacity, phenomenological, and biased noise models, we show that UIUF significantly outperforms UF, reducing the logical error rate by over an order of magnitude (at around $10^{-5}$). Moreover, UIUF achieves lower logical error rates than the Minimum Weight Perfect Matching (MWPM) decoder on rotated surface codes under both the code capacity and phenomenological noise models, while preserving efficient linear-time complexity. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:30:14 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Lin",
"Tzu-Hao",
""
],
[
"Lai",
"Ching-Yi",
""
]
] |
2506.14746 | Aldo Pacchiano | Nataly Brukhim, Aldo Pacchiano, Miroslav Dudik, Robert Schapire | On the Hardness of Bandit Learning | 13 main pages | null | null | null | cs.LG stat.ML | http://creativecommons.org/licenses/by/4.0/ | We study the task of bandit learning, also known as best-arm identification, under the assumption that the true reward function f belongs to a known, but arbitrary, function class F. We seek a general theory of bandit learnability, akin to the PAC framework for classification. Our investigation is guided by the following two questions: (1) which classes F are learnable, and (2) how they are learnable. For example, in the case of binary PAC classification, learnability is fully determined by a combinatorial dimension - the VC dimension- and can be attained via a simple algorithmic principle, namely, empirical risk minimization (ERM). In contrast to classical learning-theoretic results, our findings reveal limitations of learning in structured bandits, offering insights into the boundaries of bandit learnability. First, for the question of "which", we show that the paradigm of identifying the learnable classes via a dimension-like quantity fails for bandit learning. We give a simple proof demonstrating that no combinatorial dimension can characterize bandit learnability, even in finite classes, following a standard definition of dimension introduced by Ben-David et al. (2019). For the question of "how", we prove a computational hardness result: we construct a reward function class for which at most two queries are needed to find the optimal action, yet no algorithm can do so in polynomial time unless RP=NP. We also prove that this class admits efficient algorithms for standard algorithmic operations often considered in learning theory, such as an ERM. This implies that computational hardness is in this case inherent to the task of bandit learning. Beyond these results, we investigate additional themes such as learning under noise, trade-offs between noise models, and the relationship between query complexity and regret minimization. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:35:25 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Brukhim",
"Nataly",
""
],
[
"Pacchiano",
"Aldo",
""
],
[
"Dudik",
"Miroslav",
""
],
[
"Schapire",
"Robert",
""
]
] |
2506.14747 | Gaia De Paciani | Gaia De Paciani and Lukas Homeier and Jad C. Halimeh and Monika Aidelsburger and Fabian Grusdt | Quantum simulation of fermionic non-Abelian lattice gauge theories in $(2+1)$D with built-in gauge protection | null | null | null | null | cond-mat.quant-gas quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Recent advancements in the field of quantum simulation have significantly expanded the potential for applications, particularly in the context of lattice gauge theories (LGTs). Maintaining gauge invariance throughout a simulation remains a central challenge, especially for large-scale non-Abelian LGTs with dynamical matter, which are particularly complex in terms of engineering for experiments. Gauge-symmetry breaking is inevitable in established rishon-based schemes for alkaline-earth-like atoms (AELAs) and controlling the magnitude of its effect is an open challenge. Here, we first construct a minimal model to quantum simulate non-Abelian LGTs ensuring that the gauge constraints are met and explicitly derive their unambiguous non-Abelian nature. Second, we present a proposal for a novel gauge protection scheme using native interactions in AELAs enabling the simulation of toy models of non-Abelian $U(2)$ LGTs with dynamical fermionic matter in $(2+1)$ dimensions on large scales. Due to the simplicity of the gauge protection mechanism, based on a Zeeman shift in combination with superexchange interactions, our scheme can be naturally included in other rishon-based quantum simulation protocols. Third, we extend our approach to a fully scalable, hybrid digital-analog simulator for $U(N)$ LGTs based on Rydberg AELA with variable rishon number. The proposed general mechanism for gauge protection provides a promising path towards the long-awaited simulation of non-Abelian LGTs relevant to particle physics. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:37:40 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"De Paciani",
"Gaia",
""
],
[
"Homeier",
"Lukas",
""
],
[
"Halimeh",
"Jad C.",
""
],
[
"Aidelsburger",
"Monika",
""
],
[
"Grusdt",
"Fabian",
""
]
] |
2506.14748 | Zhuo Wu | Jaehoon Kim, Hong Liu, Chong Shangguan, Guanghui Wang, Zhuo Wu, Yisai Xue | Stability with minuscule structure for chromatic thresholds | 26 pages, 2 figures | null | null | null | math.CO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The chromatic threshold $\delta_\chi(H)$ of a graph $H$ is the infimum of $d>0$ such that the chromatic number of every $n$-vertex $H$-free graph with minimum degree at least $d n$ is bounded by a constant depending only on $H$ and $d$. Allen, B{\"o}ttcher, Griffiths, Kohayakawa, and Morris determined the chromatic threshold for every $H$; in particular, they showed that if $\chi(H)=r\ge 3$, then $\delta_\chi(H) \in\{\frac{r-3}{r-2},~\frac{2 r-5}{2 r-3},~\frac{r-2}{r-1}\}$. While the chromatic thresholds have been completely determined, rather surprisingly the structural behaviors of extremal graphs near the threshold remain unexplored.
In this paper, we establish the stability theorems for chromatic threshold problems. We prove that every $n$-vertex $H$-free graph $G$ with $\delta(G)\ge (\delta_\chi(H)-o(1))n$ and $\chi(G)=\omega(1)$ must be structurally close to one of the extremal configurations. Furthermore, we give a stronger stability result when $H$ is a clique, showing that $G$ admits a partition into independent sets and a small subgraph on sublinear number of vertices. We show that this small subgraph has fractional chromatic number $2+o(1)$ and is homomorphic to a Kneser graph defined by subsets of a logarithmic size set; both these two bounds are best possible. This is the first stability result that captures the lower-order structural features of extremal graphs.
We also study two variations of chromatic thresholds. Replacing chromatic number by its fractional counterpart, we determine the fractional chromatic thresholds for all graphs. Another variation is the bounded-VC chromatic thresholds, which was introduced by Liu, Shangguan, Skokan, and Xu very recently. Extending work of {\L}uczak and Thomass{\'e} on the triangle case, we determine the bounded-VC chromatic thresholds for all cliques. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:39:14 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Kim",
"Jaehoon",
""
],
[
"Liu",
"Hong",
""
],
[
"Shangguan",
"Chong",
""
],
[
"Wang",
"Guanghui",
""
],
[
"Wu",
"Zhuo",
""
],
[
"Xue",
"Yisai",
""
]
] |
2506.14749 | Shiyu Cheng | Shiyu Cheng, Luyao Niu, Bhaskar Ramasubramanian, Andrew Clark, Radha Poovendran | Swarm-STL: A Framework for Motion Planning in Large-Scale, Multi-Swarm Systems | null | null | null | null | eess.SY cs.SY | http://creativecommons.org/licenses/by/4.0/ | In multi-agent systems, signal temporal logic (STL) is widely used for path planning to accomplish complex objectives with formal safety guarantees. However, as the number of agents increases, existing approaches encounter significant computational challenges. Recognizing that many complex tasks require cooperation among multiple agents, we propose swarm STL specifications to describe the collective tasks that need to be achieved by a team of agents. Next, we address the motion planning problem for all the agents in two stages. First, we abstract a group of cooperating agents as a swarm and construct a reduced-dimension state space whose dimension does not increase with the number of agents. The path planning is performed at the swarm level, ensuring the safety and swarm STL specifications are satisfied. Then, we design low-level control strategies for agents within each swarm based on the path synthesized in the first step. The trajectories of agents generated by the two-step policy ensure satisfaction of the STL specifications. We evaluate our two-stage approach in both single-swarm and multi-swarm scenarios. The results demonstrate that all tasks are completed with safety guarantees. Compared to the baseline multi-agent planning approach, our method maintains computational efficiency as the number of agents increases, since the computational time scales with the number of swarms rather than the number of agents. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:40:12 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Cheng",
"Shiyu",
""
],
[
"Niu",
"Luyao",
""
],
[
"Ramasubramanian",
"Bhaskar",
""
],
[
"Clark",
"Andrew",
""
],
[
"Poovendran",
"Radha",
""
]
] |
2506.14750 | GaoBin Yang | Gaobin Yang, Maokui He, Shutong Niu, Ruoyu Wang, Hang Chen, Jun Du | Exploring Speaker Diarization with Mixture of Experts | null | null | null | null | cs.SD cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper, we propose a novel neural speaker diarization system using memory-aware multi-speaker embedding with sequence-to-sequence architecture (NSD-MS2S), which integrates a memory-aware multi-speaker embedding module with a sequence-to-sequence architecture. The system leverages a memory module to enhance speaker embeddings and employs a Seq2Seq framework to efficiently map acoustic features to speaker labels. Additionally, we explore the application of mixture of experts in speaker diarization, and introduce a Shared and Soft Mixture of Experts (SS-MoE) module to further mitigate model bias and enhance performance. Incorporating SS-MoE leads to the extended model NSD-MS2S-SSMoE. Experiments on multiple complex acoustic datasets, including CHiME-6, DiPCo, Mixer 6 and DIHARD-III evaluation sets, demonstrate meaningful improvements in robustness and generalization. The proposed methods achieve state-of-the-art results, showcasing their effectiveness in challenging real-world scenarios. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:42:54 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Yang",
"Gaobin",
""
],
[
"He",
"Maokui",
""
],
[
"Niu",
"Shutong",
""
],
[
"Wang",
"Ruoyu",
""
],
[
"Chen",
"Hang",
""
],
[
"Du",
"Jun",
""
]
] |
2506.14751 | Fugen Hagihara | Fugen Hagihara and Akitoshi Kawamura | The Ultimate Signs of Second-Order Holonomic Sequences | 26 pages, 1 figure, a full version of the same-name paper accepted to ICALP 2025 | null | 10.4230/LIPIcs.ICALP.2025.155 | null | cs.DM | http://creativecommons.org/licenses/by/4.0/ | A real-valued sequence $f = \{ f(n) \}_{n \in \mathbb{N}}$ is said to be second-order holonomic if it satisfies a linear recurrence $f (n + 2) = P (n) f (n + 1) + Q (n) f (n)$ for all sufficiently large $n$, where $P, Q \in \mathbb{R}(x)$ are rational functions. We study the ultimate sign of such a sequence, i.e., the repeated pattern that the signs of $f (n)$ follow for sufficiently large $n$. For each $P$, $Q$ we determine all the ultimate signs that $f$ can have, and show how they partition the space of initial values of $f$. This completes the prior work by Neumann, Ouaknine and Worrell, who have settled some restricted cases. As a corollary, it follows that when $P$, $Q$ have rational coefficients, $f$ either has an ultimate sign of length $1$, $2$, $3$, $4$, $6$, $8$ or $12$, or never falls into a repeated sign pattern. We also give a partial algorithm that finds the ultimate sign of $f$ (or tells that there is none) in almost all cases. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:43:13 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Hagihara",
"Fugen",
""
],
[
"Kawamura",
"Akitoshi",
""
]
] |
2506.14752 | Chase Hatcher | Chase Hatcher, Adrienne Traxler, Lily Donis, Madison Swirtz, Camila Manni, Justin Gutzwa, Charles Henderson, Ram\'on Barthelemy | Unique support from identity-based groups: Professional networks of women and LGBTQ+ physicists analyzed and compared by career sector | null | null | null | null | physics.ed-ph | http://creativecommons.org/licenses/by/4.0/ | This paper presents a social network analysis of the professional support networks of 100 LGBTQ+ and/or women PhD physicists, comparing the networks based on the career sectors of academia, industry, and government/nonprofit. The methods for constructing and analyzing the ego networks, which are novel in many ways, are explained in greater detail in an earlier publication (Hatcher et. al., 2025). We use statistical tests of independence to explore differences between sectors in terms of whole network metrics, network composition based on alter characteristics, and support types. We find that alters associated with groups (like affinity groups and personal and professional interest groups) are more likely to provide identity-based and community building support, participants in Academia have fewer personal friends in their networks while those in Industry have more, participants in Government report less instrumental support, and those in Academia report less material support. These results and others lead to suggestions for employers in these sectors on how to better support these physicists, including continuing to promote participation in affinity and interest groups, providing more material support and/or personal time in the academic sector, and more instrumental support in the form of professional development or training in the government sector. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:44:09 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Hatcher",
"Chase",
""
],
[
"Traxler",
"Adrienne",
""
],
[
"Donis",
"Lily",
""
],
[
"Swirtz",
"Madison",
""
],
[
"Manni",
"Camila",
""
],
[
"Gutzwa",
"Justin",
""
],
[
"Henderson",
"Charles",
""
],
[
"Barthelemy",
"Ramón",
""
]
] |
2506.14753 | Patsorn Sangkloy | Qinchan (Wing) Li, Kenneth Chen, Changyue (Tina) Su, Wittawat Jitkrittum, Qi Sun, Patsorn Sangkloy | Cost-Aware Routing for Efficient Text-To-Image Generation | null | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion models are well known for their ability to generate a high-fidelity image for an input prompt through an iterative denoising process. Unfortunately, the high fidelity also comes at a high computational cost due the inherently sequential generative process. In this work, we seek to optimally balance quality and computational cost, and propose a framework to allow the amount of computation to vary for each prompt, depending on its complexity. Each prompt is automatically routed to the most appropriate text-to-image generation function, which may correspond to a distinct number of denoising steps of a diffusion model, or a disparate, independent text-to-image model. Unlike uniform cost reduction techniques (e.g., distillation, model quantization), our approach achieves the optimal trade-off by learning to reserve expensive choices (e.g., 100+ denoising steps) only for a few complex prompts, and employ more economical choices (e.g., small distilled model) for less sophisticated prompts. We empirically demonstrate on COCO and DiffusionDB that by learning to route to nine already-trained text-to-image models, our approach is able to deliver an average quality that is higher than that achievable by any of these models alone. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:48:50 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Qinchan",
"",
"",
"Wing"
],
[
"Li",
"",
"",
"Tina"
],
[
"Chen",
"Kenneth",
"",
"Tina"
],
[
"Changyue",
"",
"",
"Tina"
],
[
"Su",
"",
""
],
[
"Jitkrittum",
"Wittawat",
""
],
[
"Sun",
"Qi",
""
],
[
"Sangkloy",
"Patsorn",
""
]
] |
2506.14754 | Carolina Higuera | Carolina Higuera, Akash Sharma, Taosha Fan, Chaithanya Krishna Bodduluri, Byron Boots, Michael Kaess, Mike Lambeta, Tingfan Wu, Zixi Liu, Francois Robert Hogan, Mustafa Mukadam | Tactile Beyond Pixels: Multisensory Touch Representations for Robot Manipulation | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present Sparsh-X, the first multisensory touch representations across four tactile modalities: image, audio, motion, and pressure. Trained on ~1M contact-rich interactions collected with the Digit 360 sensor, Sparsh-X captures complementary touch signals at diverse temporal and spatial scales. By leveraging self-supervised learning, Sparsh-X fuses these modalities into a unified representation that captures physical properties useful for robot manipulation tasks. We study how to effectively integrate real-world touch representations for both imitation learning and tactile adaptation of sim-trained policies, showing that Sparsh-X boosts policy success rates by 63% over an end-to-end model using tactile images and improves robustness by 90% in recovering object states from touch. Finally, we benchmark Sparsh-X ability to make inferences about physical properties, such as object-action identification, material-quantity estimation, and force estimation. Sparsh-X improves accuracy in characterizing physical properties by 48% compared to end-to-end approaches, demonstrating the advantages of multisensory pretraining for capturing features essential for dexterous manipulation. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:49:16 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Higuera",
"Carolina",
""
],
[
"Sharma",
"Akash",
""
],
[
"Fan",
"Taosha",
""
],
[
"Bodduluri",
"Chaithanya Krishna",
""
],
[
"Boots",
"Byron",
""
],
[
"Kaess",
"Michael",
""
],
[
"Lambeta",
"Mike",
""
],
[
"Wu",
"Tingfan",
""
],
[
"Liu",
"Zixi",
""
],
[
"Hogan",
"Francois Robert",
""
],
[
"Mukadam",
"Mustafa",
""
]
] |
2506.14755 | Zhengxiang Cheng | Zhengxiang Cheng, Dongping Chen, Mingyang Fu, Tianyi Zhou | Optimizing Length Compression in Large Reasoning Models | 16 pages, 7 figures, 4 tables | null | null | null | cs.AI cs.CL | http://creativecommons.org/licenses/by/4.0/ | Large Reasoning Models (LRMs) have achieved remarkable success, yet they often suffer from producing unnecessary and verbose reasoning chains. We identify a core aspect of this issue as "invalid thinking" -- models tend to repeatedly double-check their work after having derived the correct answer. To address this specific inefficiency, we move beyond the general principles of Efficacy and Efficiency to propose two new, fine-grained principles: Brevity, which advocates for eliminating redundancy, and Sufficiency, which ensures critical reasoning steps are preserved. Guided by these principles, we introduce LC-R1, a post-training method based on Group Relative Policy Optimization (GRPO). LC-R1 employs a novel combination of a Length Reward for overall conciseness and a Compress Reward that is specifically designed to remove the invalid portion of the thinking process. Extensive experiments on multiple reasoning benchmarks demonstrate that LC-R1 achieves a significant reduction in sequence length (~50%) with only a marginal (~2%) drop in accuracy, achieving a favorable trade-off point on the Pareto frontier that prioritizes high compression. Our analysis further validates the robustness of LC-R1 and provides valuable insights for developing more powerful yet computationally efficient LRMs. Our code is released at https://github.com/zxiangx/LC-R1. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:50:16 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Cheng",
"Zhengxiang",
""
],
[
"Chen",
"Dongping",
""
],
[
"Fu",
"Mingyang",
""
],
[
"Zhou",
"Tianyi",
""
]
] |
2506.14756 | Charles Wiame | Charles Wiame, Ken R. Duffy, Muriel M\'edard | Joint Error Correction and Fading Channel Estimation Enhancement Leveraging GRAND | null | null | null | null | cs.IT eess.SP math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We present a novel method for error correction in the presence of fading channel estimation errors (CEE). When such errors are significant, considerable performance losses can be observed if the wireless transceiver is not adapted. Instead of refining the estimate by increasing the pilot sequence length or improving the estimation algorithm, we propose two new approaches based on Guessing Random Additive Noise Decoding (GRAND) decoders. The first method involves testing multiple candidates for the channel estimate located in the complex neighborhood around the original pilot-based estimate. All these candidates are employed in parallel to compute log-likelihood ratios (LLR). These LLRs are used as soft input to Ordered Reliability Bits GRAND (ORBGRAND). Posterior likelihood formulas associated with ORBGRAND are then computed to determine which channel candidate leads to the most probable codeword. The second method is a refined version of the first approach accounting for the presence of residual CEE in the LLR computation. The performance of these two techniques is evaluated for [128,112] 5G NR CA-Polar and CRC codes. For the considered settings, block error rate (BLER) gains of several dBs are observed compared to cases where CEE is ignored. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:50:44 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Wiame",
"Charles",
""
],
[
"Duffy",
"Ken R.",
""
],
[
"Médard",
"Muriel",
""
]
] |
2506.14757 | Daniel Schiller | Caroline Heneka, Florian Nieser, Ayodele Ore, Tilman Plehn, Daniel Schiller | Large Language Models -- the Future of Fundamental Physics? | 33 pages, 9 figures | null | null | null | astro-ph.CO astro-ph.IM hep-ph physics.data-an | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | For many fundamental physics applications, transformers, as the state of the art in learning complex correlations, benefit from pretraining on quasi-out-of-domain data. The obvious question is whether we can exploit Large Language Models, requiring proper out-of-domain transfer learning. We show how the Qwen2.5 LLM can be used to analyze and generate SKA data, specifically 3D maps of the cosmological large-scale structure for a large part of the observable Universe. We combine the LLM with connector networks and show, for cosmological parameter regression and lightcone generation, that this Lightcone LLM (L3M) with Qwen2.5 weights outperforms standard initialization and compares favorably with dedicated networks of matching size. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:51:19 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Heneka",
"Caroline",
""
],
[
"Nieser",
"Florian",
""
],
[
"Ore",
"Ayodele",
""
],
[
"Plehn",
"Tilman",
""
],
[
"Schiller",
"Daniel",
""
]
] |
2506.14758 | Daixuan Cheng | Daixuan Cheng, Shaohan Huang, Xuekai Zhu, Bo Dai, Wayne Xin Zhao, Zhenliang Zhang, Furu Wei | Reasoning with Exploration: An Entropy Perspective | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Balancing exploration and exploitation is a central goal in reinforcement learning (RL). Despite recent advances in enhancing language model (LM) reasoning, most methods lean toward exploitation, and increasingly encounter performance plateaus. In this work, we revisit entropy -- a signal of exploration in RL -- and examine its relationship to exploratory reasoning in LMs. Through empirical analysis, we uncover strong positive correlations between high-entropy regions and three types of exploratory reasoning actions: (1) pivotal tokens that determine or connect logical steps, (2) reflective actions such as self-verification and correction, and (3) rare behaviors under-explored by the base LMs. Motivated by this, we introduce a minimal modification to standard RL with only one line of code: augmenting the advantage function with an entropy-based term. Unlike traditional maximum-entropy methods which encourage exploration by promoting uncertainty, we encourage exploration by promoting longer and deeper reasoning chains. Notably, our method achieves significant gains on the Pass@K metric -- an upper-bound estimator of LM reasoning capabilities -- even when evaluated with extremely large K values, pushing the boundaries of LM reasoning. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:54:03 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Cheng",
"Daixuan",
""
],
[
"Huang",
"Shaohan",
""
],
[
"Zhu",
"Xuekai",
""
],
[
"Dai",
"Bo",
""
],
[
"Zhao",
"Wayne Xin",
""
],
[
"Zhang",
"Zhenliang",
""
],
[
"Wei",
"Furu",
""
]
] |
2506.14759 | Patrick Del Vecchio | Patrick Del Vecchio, Stefano Bosco, Daniel Loss and Oussama Moutanabbir | Fully Tunable Strong Spin-Orbit Interactions in Light Hole Germanium Quantum Channels | null | null | null | null | cond-mat.mes-hall | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Spin-orbit interaction (SOI) is a fundamental component for electrically driven spin qubits and hybrid superconducting-semiconducting systems. In particular, Rashba SOI (RSOI) is a key mechanism enabling all-electrical spin manipulation schemes. However, in common planar systems, RSOI is weak because of the small mixing between heavy holes (HH) and light holes (LH), and instead relies on complex strain and interface phenomena that are hard to reliably harness in experiment. Here, MOS-like epitaxial Ge on relaxed \GeSn{} is introduced and shown to exhibit an inherently large, highly gate-tunable RSOI that is compatible with both spin qubits and hybrid devices. This large RSOI is a consequence of the LH-like ground state in Ge. Notably, the built-in asymmetry of the device causes the RSOI to completely vanish at specific gate fields, effectively acting as an on/off SOI switch. The LH $g$-tensor is less anisotropic than that of state-of-the-art HH qubits, alleviating precise magnetic field orientation requirements. The large in-plane $g$-factor also facilitates the integration of superconductors. Moreover, the out-of-plane $g$-factor is strongly gate-tunable and completely vanishes at specific gate fields. Thus, this material system combines the large RSOI with the scalability of planar devices, paving the way towards robust spin qubit applications and enabling access to new regimes of complex spin physics. | [
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:54:29 GMT"
}
] | 2025-06-18T00:00:00 | [
[
"Del Vecchio",
"Patrick",
""
],
[
"Bosco",
"Stefano",
""
],
[
"Loss",
"Daniel",
""
],
[
"Moutanabbir",
"Oussama",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.