publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2024-03-14T05:34:35Z
|
http://arxiv.org/pdf/2403.09750v1
|
http://arxiv.org/abs/2403.09750v1
|
Zhuoqun Li, Hongyu Lin, Yaojie Lu, Hao Xiang, Xianpei Han, Le Sun
|
2024-03-14T05:34:35Z
|
cs.CL, cs.AI
|
Declarative knowledge and procedural knowledge are two key parts in
meta-cognitive theory, and these two hold significant importance in
pre-training and inference of LLMs. However, a comprehensive analysis comparing
these two types of knowledge is lacking, primarily due to challenges in
definition, probing and quantitative assessment. In this paper, we explore from
a new perspective by providing ground-truth knowledge for LLMs and evaluating
the effective score. Through extensive experiments with widely-used datasets
and models, we get conclusions: (1) In most tasks, benefits from declarative
knowledge are greater than those from procedural knowledge. (2) Profits of
procedural knowledge are larger than declarative knowledge only in reasoning
tasks with simple logic. (3) As pre-training progresses and size increases,
model ability to utilize both kinds of knowledge significantly improves, but in
different speed. We do detailed analysis for the findings and this can provide
primary guidance for evaluation and enhancement of large language models.
|
cs.CL
|
Meta-Cognitive Analysis: Evaluating Declarative and Procedural Knowledge in Datasets and Large Language Models
|
2024-03-14T05:29:35Z
|
http://arxiv.org/pdf/2403.09113v2
|
http://arxiv.org/abs/2403.09113v2
|
Ruiyi Zhang, Rushi Qiang, Sai Ashish Somayajula, Pengtao Xie
|
2024-03-17T17:55:47Z
|
cs.CL, cs.AI, cs.LG
|
Large-scale pretraining followed by task-specific finetuning has achieved
great success in various NLP tasks. Since finetuning all parameters of large
pretrained models poses substantial computational and memory challenges,
several efficient finetuning methods have been developed. Among them, low-rank
adaptation (LoRA), which finetunes low-rank incremental update matrices on top
of frozen pretrained weights, has proven particularly effective. Nonetheless,
LoRA's uniform rank assignment across all layers, along with its reliance on an
exhaustive search to find the best rank, leads to high computation costs and
suboptimal finetuning performance. To address these limitations, we introduce
AutoLoRA, a meta learning based framework for automatically identifying the
optimal rank of each LoRA layer. AutoLoRA associates each rank-1 matrix in a
low-rank update matrix with a selection variable, which determines whether the
rank-1 matrix should be discarded. A meta learning based method is developed to
learn these selection variables. The optimal rank is determined by thresholding
the values of these variables. Our comprehensive experiments on natural
language understanding, generation, and sequence labeling demonstrate the
effectiveness of AutoLoRA.
|
cs.CL
|
AutoLoRA: Automatically Tuning Matrix Ranks in Low-Rank Adaptation Based on Meta Learning
|
2024-03-14T04:43:02Z
|
http://arxiv.org/pdf/2403.09097v1
|
http://arxiv.org/abs/2403.09097v1
|
Autumn Toney-Wails, Christian Schoeberl, James Dunham
|
2024-03-14T04:43:02Z
|
cs.CL
|
Identifying scientific publications that are within a dynamic field of
research often requires costly annotation by subject-matter experts. Resources
like widely-accepted classification criteria or field taxonomies are
unavailable for a domain like artificial intelligence (AI), which spans
emerging topics and technologies. We address these challenges by inferring a
functional definition of AI research from existing expert labels, and then
evaluating state-of-the-art chatbot models on the task of expert data
annotation. Using the arXiv publication database as ground-truth, we experiment
with prompt engineering for GPT chatbot models to identify an alternative,
automated expert annotation pipeline that assigns AI labels with 94% accuracy.
For comparison, we fine-tune SPECTER, a transformer language model pre-trained
on scientific publications, that achieves 96% accuracy (only 2% higher than
GPT) on classifying AI publications. Our results indicate that with effective
prompt engineering, chatbots can be used as reliable data annotators even where
subject-area expertise is required. To evaluate the utility of
chatbot-annotated datasets on downstream classification tasks, we train a new
classifier on GPT-labeled data and compare its performance to the arXiv-trained
model. The classifier trained on GPT-labeled data outperforms the arXiv-trained
model by nine percentage points, achieving 82% accuracy.
|
cs.CL
|
AI on AI: Exploring the Utility of GPT as an Expert Annotator of AI Publications
|
2024-03-14T04:32:13Z
|
http://arxiv.org/pdf/2403.09092v1
|
http://arxiv.org/abs/2403.09092v1
|
Yupeng Li, Haorui He, Jin Bai, Dacheng Wen
|
2024-03-14T04:32:13Z
|
cs.CL, cs.AI
|
The prevalence of fake news across various online sources has had a
significant influence on the public. Existing Chinese fake news detection
datasets are limited to news sourced solely from Weibo. However, fake news
originating from multiple sources exhibits diversity in various aspects,
including its content and social context. Methods trained on purely one single
news source can hardly be applicable to real-world scenarios. Our pilot
experiment demonstrates that the F1 score of the state-of-the-art method that
learns from a large Chinese fake news detection dataset, Weibo-21, drops
significantly from 0.943 to 0.470 when the test data is changed to multi-source
news data, failing to identify more than one-third of the multi-source fake
news. To address this limitation, we constructed the first multi-source
benchmark dataset for Chinese fake news detection, termed MCFEND, which is
composed of news we collected from diverse sources such as social platforms,
messaging apps, and traditional online news outlets. Notably, such news has
been fact-checked by 14 authoritative fact-checking agencies worldwide. In
addition, various existing Chinese fake news detection methods are thoroughly
evaluated on our proposed dataset in cross-source, multi-source, and unseen
source ways. MCFEND, as a benchmark dataset, aims to advance Chinese fake news
detection approaches in real-world scenarios.
|
cs.CL
|
MCFEND: A Multi-source Benchmark Dataset for Chinese Fake News Detection
|
2024-03-14T04:06:13Z
|
http://arxiv.org/pdf/2403.09085v1
|
http://arxiv.org/abs/2403.09085v1
|
Kai Xiong, Xiao Ding, Ting Liu, Bing Qin, Dongliang Xu, Qing Yang, Hongtao Liu, Yixin Cao
|
2024-03-14T04:06:13Z
|
cs.CL, cs.AI
|
Large language models (LLMs) have developed impressive performance and strong
explainability across various reasoning scenarios, marking a significant stride
towards mimicking human-like intelligence. Despite this, when tasked with
simple questions supported by a generic fact, LLMs often fail to provide
consistent and precise answers, indicating a deficiency in abstract reasoning
abilities. This has sparked a vigorous debate about whether LLMs are genuinely
reasoning or merely memorizing. In light of this, we design a preliminary study
to quantify and delve into the abstract reasoning abilities of existing LLMs.
Our findings reveal a substantial discrepancy between their general reasoning
and abstract reasoning performances. To relieve this problem, we tailor an
abstract reasoning dataset (AbsR) together with a meaningful learning paradigm
to teach LLMs how to leverage generic facts for reasoning purposes. The results
show that our approach not only boosts the general reasoning performance of
LLMs but also makes considerable strides towards their capacity for abstract
reasoning, moving beyond simple memorization or imitation to a more nuanced
understanding and application of generic facts.
|
cs.CL
|
Meaningful Learning: Advancing Abstract Reasoning in Large Language Models via Generic Fact Guidance
|
2024-03-14T03:49:36Z
|
http://arxiv.org/pdf/2403.09077v1
|
http://arxiv.org/abs/2403.09077v1
|
Abuzar Royesh, Olamide Oladeji
|
2024-03-14T03:49:36Z
|
cs.CL
|
Despite the need for financial data on company activities in developing
countries for development research and economic analysis, such data does not
exist. In this project, we develop and evaluate two Natural Language Processing
(NLP) based techniques to address this issue. First, we curate a custom dataset
specific to the domain of financial text data on developing countries and
explore multiple approaches for information extraction. We then explore a
text-to-text approach with the transformer-based T5 model with the goal of
undertaking simultaneous NER and relation extraction. We find that this model
is able to learn the custom text structure output data corresponding to the
entities and their relations, resulting in an accuracy of 92.44\%, a precision
of 68.25\% and a recall of 54.20\% from our best T5 model on the combined task.
Secondly, we explore an approach with sequential NER and relation extration.
For the NER, we run pre-trained and fine-tuned models using SpaCy, and we
develop a custom relation extraction model using SpaCy's Dependency Parser
output and some heuristics to determine entity relationships \cite{spacy}. We
obtain an accuracy of 84.72\%, a precision of 6.06\% and a recall of 5.57\% on
this sequential task.
|
cs.CL
|
Information Extraction: An application to the domain of hyper-local financial data on developing countries
|
2024-03-14T03:33:46Z
|
http://arxiv.org/pdf/2403.09073v1
|
http://arxiv.org/abs/2403.09073v1
|
Yongyu Mu, Peinan Feng, Zhiquan Cao, Yuzhang Wu, Bei Li, Chenglong Wang, Tong Xiao, Kai Song, Tongran Liu, Chunliang Zhang, Jingbo Zhu
|
2024-03-14T03:33:46Z
|
cs.CL
|
In this study, we reveal an in-context learning (ICL) capability of
multilingual large language models (LLMs): by translating the input to several
languages, we provide Parallel Input in Multiple Languages (PiM) to LLMs, which
significantly enhances their comprehension abilities. To test this capability,
we design extensive experiments encompassing 8 typical datasets, 7 languages
and 8 state-of-the-art multilingual LLMs. Experimental results show that (1)
incorporating more languages help PiM surpass the conventional ICL further; (2)
even combining with the translations that are inferior to baseline performance
can also help. Moreover, by examining the activated neurons in LLMs, we
discover a counterintuitive but interesting phenomenon. Contrary to the common
thought that PiM would activate more neurons than monolingual input to leverage
knowledge learned from diverse languages, PiM actually inhibits neurons and
promotes more precise neuron activation especially when more languages are
added. This phenomenon aligns with the neuroscience insight about synaptic
pruning, which removes less used neural connections, strengthens remainders,
and then enhances brain intelligence.
|
cs.CL
|
Large Language Models are Parallel Multilingual Learners
|
2024-03-14T03:29:58Z
|
http://arxiv.org/pdf/2403.09072v1
|
http://arxiv.org/abs/2403.09072v1
|
Sipeng Zheng, Bohan Zhou, Yicheng Feng, Ye Wang, Zongqing Lu
|
2024-03-14T03:29:58Z
|
cs.CV, cs.AI, cs.CL
|
In this paper, we propose \textbf{UniCode}, a novel approach within the
domain of multimodal large language models (MLLMs) that learns a unified
codebook to efficiently tokenize visual, text, and potentially other types of
signals. This innovation addresses a critical limitation in existing MLLMs:
their reliance on a text-only codebook, which restricts MLLM's ability to
generate images and texts in a multimodal context. Towards this end, we propose
a language-driven iterative training paradigm, coupled with an in-context
pre-training task we term ``image decompression'', enabling our model to
interpret compressed visual data and generate high-quality images.The unified
codebook empowers our model to extend visual instruction tuning to
non-linguistic generation tasks. Moreover, UniCode is adaptable to diverse
stacked quantization approaches in order to compress visual signals into a more
compact token representation. Despite using significantly fewer parameters and
less data during training, Unicode demonstrates promising capabilities in
visual reconstruction and generation. It also achieves performances comparable
to leading MLLMs across a spectrum of VQA benchmarks.
|
cs.CV
|
UniCode: Learning a Unified Codebook for Multimodal Large Language Models
|
2024-03-14T02:56:38Z
|
http://arxiv.org/pdf/2403.09059v1
|
http://arxiv.org/abs/2403.09059v1
|
Pasquale Balsebre, Weiming Huang, Gao Cong
|
2024-03-14T02:56:38Z
|
cs.CL
|
Large Language Models (LLMs) are poised to play an increasingly important
role in our lives, providing assistance across a wide array of tasks. In the
geospatial domain, LLMs have demonstrated the ability to answer generic
questions, such as identifying a country's capital; nonetheless, their utility
is hindered when it comes to answering fine-grained questions about specific
places, such as grocery stores or restaurants, which constitute essential
aspects of people's everyday lives. This is mainly because the places in our
cities haven't been systematically fed into LLMs, so as to understand and
memorize them. This study introduces a novel framework for fine-tuning a
pre-trained model on city-specific data, to enable it to provide accurate
recommendations, while minimizing hallucinations. We share our model, LAMP, and
the data used to train it. We conduct experiments to analyze its ability to
correctly retrieving spatial objects, and compare it to well-known open- and
closed- source language models, such as GPT-4. Finally, we explore its emerging
capabilities through a case study on day planning.
|
cs.CL
|
LAMP: A Language Model on the Map
|
2024-03-14T02:55:37Z
|
http://arxiv.org/pdf/2403.09057v3
|
http://arxiv.org/abs/2403.09057v3
|
Dong Yuan, Eti Rastogi, Gautam Naik, Sree Prasanna Rajagopal, Sagar Goyal, Fen Zhao, Bharath Chintagunta, Jeff Ward
|
2024-04-03T18:08:30Z
|
cs.CL, cs.AI
|
LLMs are revolutionizing NLP tasks. However, the use of the most advanced
LLMs, such as GPT-4, is often prohibitively expensive for most specialized
fields. We introduce HEAL, the first continuously trained 13B LLaMA2-based LLM
that is purpose-built for medical conversations and measured on automated
scribing. Our results demonstrate that HEAL outperforms GPT-4 and PMC-LLaMA in
PubMedQA, with an accuracy of 78.4\%. It also achieves parity with GPT-4 in
generating medical notes. Remarkably, HEAL surpasses GPT-4 and Med-PaLM 2 in
identifying more correct medical concepts and exceeds the performance of human
scribes and other comparable models in correctness and completeness.
|
cs.CL
|
A Continued Pretrained LLM Approach for Automatic Medical Note Generation
|
2024-03-14T02:42:42Z
|
http://arxiv.org/pdf/2403.09054v2
|
http://arxiv.org/abs/2403.09054v2
|
Muhammad Adnan, Akhil Arunkumar, Gaurav Jain, Prashant J. Nair, Ilya Soloveychik, Purushotham Kamath
|
2024-04-06T00:22:37Z
|
cs.LG, cs.AI, cs.AR, cs.CL, 68U35, I.2.7; C.0
|
Transformers have emerged as the underpinning architecture for Large Language
Models (LLMs). In generative language models, the inference process involves
two primary phases: prompt processing and token generation. Token generation,
which constitutes the majority of the computational workload, primarily entails
vector-matrix multiplications and interactions with the Key-Value (KV) Cache.
This phase is constrained by memory bandwidth due to the overhead of
transferring weights and KV cache values from the memory system to the
computing units. This memory bottleneck becomes particularly pronounced in
applications that require long-context and extensive text generation, both of
which are increasingly crucial for LLMs.
This paper introduces "Keyformer", an innovative inference-time approach, to
mitigate the challenges associated with KV cache size and memory bandwidth
utilization. Keyformer leverages the observation that approximately 90% of the
attention weight in generative inference focuses on a specific subset of
tokens, referred to as "key" tokens. Keyformer retains only the key tokens in
the KV cache by identifying these crucial tokens using a novel score function.
This approach effectively reduces both the KV cache size and memory bandwidth
usage without compromising model accuracy. We evaluate Keyformer's performance
across three foundational models: GPT-J, Cerebras-GPT, and MPT, which employ
various positional embedding algorithms. Our assessment encompasses a variety
of tasks, with a particular emphasis on summarization and conversation tasks
involving extended contexts. Keyformer's reduction of KV cache reduces
inference latency by 2.1x and improves token generation throughput by 2.4x,
while preserving the model's accuracy.
|
cs.LG
|
Keyformer: KV Cache Reduction through Key Tokens Selection for Efficient Generative Inference
|
2024-03-14T02:26:31Z
|
http://arxiv.org/pdf/2403.09040v1
|
http://arxiv.org/abs/2403.09040v1
|
Jennifer Hsia, Afreen Shaikh, Zhiruo Wang, Graham Neubig
|
2024-03-14T02:26:31Z
|
cs.CL
|
Retrieval-augmented generation (RAG) greatly benefits language models (LMs)
by providing additional context for tasks such as document-based question
answering (DBQA). Despite its potential, the power of RAG is highly dependent
on its configuration, raising the question: What is the optimal RAG
configuration? To answer this, we introduce the RAGGED framework to analyze and
optimize RAG systems. On a set of representative DBQA tasks, we study two
classic sparse and dense retrievers, and four top-performing LMs in
encoder-decoder and decoder-only architectures. Through RAGGED, we uncover that
different models suit substantially varied RAG setups. While encoder-decoder
models monotonically improve with more documents, we find decoder-only models
can only effectively use < 5 documents, despite often having a longer context
window. RAGGED offers further insights into LMs' context utilization habits,
where we find that encoder-decoder models rely more on contexts and are thus
more sensitive to retrieval quality, while decoder-only models tend to rely on
knowledge memorized during training.
|
cs.CL
|
RAGGED: Towards Informed Design of Retrieval Augmented Generation Systems
|
2024-03-14T02:25:35Z
|
http://arxiv.org/pdf/2403.09037v1
|
http://arxiv.org/abs/2403.09037v1
|
Qinyu Zhao, Ming Xu, Kartik Gupta, Akshay Asthana, Liang Zheng, Stephen Gould
|
2024-03-14T02:25:35Z
|
cs.CV, cs.CL
|
Large vision-language models (LVLMs), designed to interpret and respond to
human instructions, occasionally generate hallucinated or harmful content due
to inappropriate instructions. This study uses linear probing to shed light on
the hidden knowledge at the output layer of LVLMs. We demonstrate that the
logit distributions of the first tokens contain sufficient information to
determine whether to respond to the instructions, including recognizing
unanswerable visual questions, defending against multi-modal jailbreaking
attack, and identifying deceptive questions. Such hidden knowledge is gradually
lost in logits of subsequent tokens during response generation. Then, we
illustrate a simple decoding strategy at the generation of the first token,
effectively improving the generated content. In experiments, we find a few
interesting insights: First, the CLIP model already contains a strong signal
for solving these tasks, indicating potential bias in the existing datasets.
Second, we observe performance improvement by utilizing the first logit
distributions on three additional tasks, including indicting uncertainty in
math solving, mitigating hallucination, and image classification. Last, with
the same training data, simply finetuning LVLMs improve models' performance but
is still inferior to linear probing on these tasks.
|
cs.CV
|
The First to Know: How Token Distributions Reveal Hidden Knowledge in Large Vision-Language Models?
|
2024-03-14T01:51:35Z
|
http://arxiv.org/pdf/2403.09032v1
|
http://arxiv.org/abs/2403.09032v1
|
Martin Weyssow, Aton Kamanda, Houari Sahraoui
|
2024-03-14T01:51:35Z
|
cs.SE, cs.CL, cs.LG
|
Evaluating the alignment of large language models (LLMs) with user-defined
coding preferences is a challenging endeavour that requires assessing intricate
textual LLMs' outputs. By relying on automated metrics and static analysis
tools, existing benchmarks fail to assess nuances in user instructions and LLM
outputs, highlighting the need for large-scale datasets and benchmarks for LLM
preference alignment. In this paper, we introduce CodeUltraFeedback, a
preference dataset of 10,000 complex instructions to tune and align LLMs to
coding preferences through AI feedback. We generate responses to the
instructions using a pool of 14 diverse LLMs, which we then annotate according
to their alignment with five coding preferences using the LLM-as-a-Judge
approach with GPT-3.5, producing both numerical and textual feedback. We also
present CODAL-Bench, a benchmark for assessing LLM alignment with these coding
preferences. Our results show that CodeLlama-7B-Instruct, aligned through
reinforcement learning from AI feedback (RLAIF) with direct preference
optimization (DPO) using CodeUltraFeedback's AI feedback data, outperforms 34B
LLMs on CODAL-Bench, validating the utility of CodeUltraFeedback for preference
tuning. Furthermore, we show our DPO-aligned CodeLlama model improves
functional correctness on HumanEval+ compared to the unaligned base model.
Therefore, our contributions bridge the gap in preference tuning of LLMs for
code and set the stage for further advancements in model alignment and RLAIF
for code intelligence. Our code and data are available at
https://github.com/martin-wey/CodeUltraFeedback.
|
cs.SE
|
CodeUltraFeedback: An LLM-as-a-Judge Dataset for Aligning Large Language Models to Coding Preferences
|
2024-03-14T01:40:23Z
|
http://arxiv.org/pdf/2403.09028v1
|
http://arxiv.org/abs/2403.09028v1
|
Ahmed Masry, Mehrad Shahmohammadi, Md Rizwan Parvez, Enamul Hoque, Shafiq Joty
|
2024-03-14T01:40:23Z
|
cs.CL
|
Charts provide visual representations of data and are widely used for
analyzing information, addressing queries, and conveying insights to others.
Various chart-related downstream tasks have emerged recently, such as
question-answering and summarization. A common strategy to solve these tasks is
to fine-tune various models originally trained on vision tasks language.
However, such task-specific models are not capable of solving a wide range of
chart-related tasks, constraining their real-world applicability. To overcome
these challenges, we introduce ChartInstruct: a novel chart-specific
vision-language Instruction-following dataset comprising 191K instructions
generated with 71K charts. We then present two distinct systems for instruction
tuning on such datasets: (1) an end-to-end model that connects a vision encoder
for chart understanding with a LLM; and (2) a pipeline model that employs a
two-step approach to extract chart data tables and input them into the LLM. In
experiments on four downstream tasks, we first show the effectiveness of our
model--achieving a new set of state-of-the-art results. Further evaluation
shows that our instruction-tuning approach supports a wide array of real-world
chart comprehension and reasoning scenarios, thereby expanding the scope and
applicability of our models to new kinds of tasks.
|
cs.CL
|
ChartInstruct: Instruction Tuning for Chart Comprehension and Reasoning
|
2024-03-14T01:28:13Z
|
http://arxiv.org/pdf/2403.09024v1
|
http://arxiv.org/abs/2403.09024v1
|
Hyunji Lee, Doyoung Kim, Jihoon Jun, Sejune Joo, Joel Jang, Kyoung-Woon On, Minjoon Seo
|
2024-03-14T01:28:13Z
|
cs.CL, cs.AI
|
In this work, we introduce a semiparametric token-sequence co-supervision
training method. It trains a language model by simultaneously leveraging
supervision from the traditional next token prediction loss which is calculated
over the parametric token embedding space and the next sequence prediction loss
which is calculated over the nonparametric sequence embedding space. The
nonparametric sequence embedding space is constructed by a separate language
model tasked to condense an input text into a single representative embedding.
Our experiments demonstrate that a model trained via both supervisions
consistently surpasses models trained via each supervision independently.
Analysis suggests that this co-supervision encourages a broader generalization
capability across the model. Especially, the robustness of parametric token
space which is established during the pretraining step tends to effectively
enhance the stability of nonparametric sequence embedding space, a new space
established by another language model.
|
cs.CL
|
Semiparametric Token-Sequence Co-Supervision
|
2024-03-14T00:45:24Z
|
http://arxiv.org/pdf/2403.09017v2
|
http://arxiv.org/abs/2403.09017v2
|
Emad A. Alghamdi, Reem I. Masoud, Deema Alnuhait, Afnan Y. Alomairi, Ahmed Ashraf, Mohamed Zaytoon
|
2024-03-15T23:52:18Z
|
cs.CL
|
The swift progress and widespread acceptance of artificial intelligence (AI)
systems highlight a pressing requirement to comprehend both the capabilities
and potential risks associated with AI. Given the linguistic complexity,
cultural richness, and underrepresented status of Arabic in AI research, there
is a pressing need to focus on Large Language Models (LLMs) performance and
safety for Arabic related tasks. Despite some progress in their development,
there is a lack of comprehensive trustworthiness evaluation benchmarks which
presents a major challenge in accurately assessing and improving the safety of
LLMs when prompted in Arabic. In this paper, we introduce AraTrust, the first
comprehensive trustworthiness benchmark for LLMs in Arabic. AraTrust comprises
516 human-written multiple-choice questions addressing diverse dimensions
related to truthfulness, ethics, safety, physical health, mental health,
unfairness, illegal activities, privacy, and offensive language. We evaluated a
set of LLMs against our benchmark to assess their trustworthiness. GPT-4 was
the most trustworthy LLM, while open-source models, particularly AceGPT 7B and
Jais 13B, struggled to achieve a score of 60% in our benchmark.
|
cs.CL
|
AraTrust: An Evaluation of Trustworthiness for LLMs in Arabic
|
2024-03-14T00:35:39Z
|
http://arxiv.org/pdf/2403.09747v1
|
http://arxiv.org/abs/2403.09747v1
|
Guanghua Li, Wensheng Lu, Wei Zhang, Defu Lian, Kezhong Lu, Rui Mao, Kai Shu, Hao Liao
|
2024-03-14T00:35:39Z
|
cs.CL, cs.AI
|
The proliferation of fake news has had far-reaching implications on politics,
the economy, and society at large. While Fake news detection methods have been
employed to mitigate this issue, they primarily depend on two essential
elements: the quality and relevance of the evidence, and the effectiveness of
the verdict prediction mechanism. Traditional methods, which often source
information from static repositories like Wikipedia, are limited by outdated or
incomplete data, particularly for emerging or rare claims. Large Language
Models (LLMs), known for their remarkable reasoning and generative
capabilities, introduce a new frontier for fake news detection. However, like
traditional methods, LLM-based solutions also grapple with the limitations of
stale and long-tail knowledge. Additionally, retrieval-enhanced LLMs frequently
struggle with issues such as low-quality evidence retrieval and context length
constraints. To address these challenges, we introduce a novel,
retrieval-augmented LLMs framework--the first of its kind to automatically and
strategically extract key evidence from web sources for claim verification.
Employing a multi-round retrieval strategy, our framework ensures the
acquisition of sufficient, relevant evidence, thereby enhancing performance.
Comprehensive experiments across three real-world datasets validate the
framework's superiority over existing methods. Importantly, our model not only
delivers accurate verdicts but also offers human-readable explanations to
improve result interpretability.
|
cs.CL
|
Re-Search for The Truth: Multi-round Retrieval-augmented Large Language Models are Strong Fake News Detectors
|
2024-03-13T23:25:30Z
|
http://arxiv.org/pdf/2403.08994v2
|
http://arxiv.org/abs/2403.08994v2
|
Lei Gao, Yue Niu, Tingting Tang, Salman Avestimehr, Murali Annavaram
|
2024-04-01T05:45:47Z
|
cs.CL
|
Language models (LMs) have greatly propelled the research on natural language
processing. However, LMs also raise concerns regarding the generation of biased
or toxic content and the potential disclosure of private information from the
training dataset. In this work, we present a new efficient approach, Ethos,
that rectifies LMs to mitigate toxicity and bias in outputs and avoid privacy
leakage. Ethos is built on task arithmetic. However, unlike current task
arithmetic algorithms, Ethos distinguishes general beneficial and undesired
knowledge when reconstructing task vectors. Specifically, Ethos first obtains a
set of principal components from the pre-trained models using singular value
decomposition. Then, by projecting the task vector onto principal components,
Ethos identifies the principal components that encode general or undesired
knowledge. Ethos performs negating using the task vector with undesired
knowledge only, thereby minimizing collateral damage on general model utility.
We demonstrate the efficacy of our approach on three different tasks:
debiasing, detoxification, and memorization unlearning. Evaluations show Ethos
is more effective in removing undesired knowledge and maintaining the overall
model performance compared to current task arithmetic methods.
|
cs.CL
|
Ethos: Rectifying Language Models in Orthogonal Parameter Space
|
2024-03-13T23:14:35Z
|
http://arxiv.org/pdf/2403.09744v1
|
http://arxiv.org/abs/2403.09744v1
|
Sven Jacobs, Steffen Jaschke
|
2024-03-13T23:14:35Z
|
cs.CL, cs.AI, cs.CY, cs.HC
|
This study investigates the application of large language models,
specifically GPT-4, to enhance programming education. The research outlines the
design of a web application that uses GPT-4 to provide feedback on programming
tasks, without giving away the solution. A web application for working on
programming tasks was developed for the study and evaluated with 51 students
over the course of one semester. The results show that most of the feedback
generated by GPT-4 effectively addressed code errors. However, challenges with
incorrect suggestions and hallucinated issues indicate the need for further
improvements.
|
cs.CL
|
Evaluating the Application of Large Language Models to Generate Feedback in Programming Education
|
2024-03-13T22:06:03Z
|
http://arxiv.org/pdf/2403.08978v1
|
http://arxiv.org/abs/2403.08978v1
|
Yao Fu, Dong-Ki Kim, Jaekyeom Kim, Sungryull Sohn, Lajanugen Logeswaran, Kyunghoon Bae, Honglak Lee
|
2024-03-13T22:06:03Z
|
cs.CL, cs.LG
|
The primary limitation of large language models (LLMs) is their restricted
understanding of the world. This poses significant difficulties for LLM-based
agents, particularly in domains where pre-trained LLMs lack sufficient
knowledge. In this paper, we introduce a novel framework, called AutoGuide,
that bridges the knowledge gap in pre-trained LLMs by leveraging implicit
knowledge in offline experiences. Specifically, AutoGuide effectively extracts
knowledge embedded in offline data by extracting a set of state-aware
guidelines. Importantly, each state-aware guideline is expressed in concise
natural language and follows a conditional structure, clearly describing the
state where it is applicable. As such, the resulting guidelines enable a
principled way to provide helpful knowledge pertinent to an agent's current
decision-making process. We show that our approach outperforms competitive
LLM-based baselines by a large margin in sequential decision-making benchmarks.
|
cs.CL
|
AutoGuide: Automated Generation and Selection of State-Aware Guidelines for Large Language Model Agents
|
2024-03-13T21:39:39Z
|
http://arxiv.org/pdf/2403.09743v1
|
http://arxiv.org/abs/2403.09743v1
|
Christian A. Schiller
|
2024-03-13T21:39:39Z
|
cs.CL, cs.AI, cs.LG
|
The launch of ChatGPT by OpenAI in November 2022 marked a pivotal moment for
Artificial Intelligence, introducing Large Language Models (LLMs) to the
mainstream and setting new records in user adoption. LLMs, particularly
ChatGPT, trained on extensive internet data, demonstrate remarkable
conversational capabilities across various domains, suggesting a significant
impact on the workforce. However, these models are susceptible to errors -
"hallucinations" and omissions, generating incorrect or incomplete information.
This poses risks especially in contexts where accuracy is crucial, such as
legal compliance, medicine or fine-grained process frameworks.
There are both technical and human solutions to cope with this isse. This
paper explores the human factors that enable users to detect errors in LLM
outputs, a critical component in mitigating risks associated with their use in
professional settings. Understanding these factors is essential for
organizations aiming to leverage LLM technology efficiently, guiding targeted
training and deployment strategies to enhance error detection by users. This
approach not only aims to optimize the use of LLMs but also to prevent
potential downstream issues stemming from reliance on inaccurate model
responses. The research emphasizes the balance between technological
advancement and human insight in maximizing the benefits of LLMs while
minimizing the risks, particularly in areas where precision is paramount.
This paper performs a systematic literature research on this research topic,
analyses and synthesizes the findings, and outlines future research directions.
Literature selection cut-off date is January 11th 2024.
|
cs.CL
|
The Human Factor in Detecting Errors of Large Language Models: A Systematic Literature Review and Future Research Directions
|
2024-03-13T20:25:27Z
|
http://arxiv.org/pdf/2403.08946v1
|
http://arxiv.org/abs/2403.08946v1
|
Xuansheng Wu, Haiyan Zhao, Yaochen Zhu, Yucheng Shi, Fan Yang, Tianming Liu, Xiaoming Zhai, Wenlin Yao, Jundong Li, Mengnan Du, Ninghao Liu
|
2024-03-13T20:25:27Z
|
cs.LG, cs.CL, cs.CY
|
Explainable AI (XAI) refers to techniques that provide human-understandable
insights into the workings of AI models. Recently, the focus of XAI is being
extended towards Large Language Models (LLMs) which are often criticized for
their lack of transparency. This extension calls for a significant
transformation in XAI methodologies because of two reasons. First, many
existing XAI methods cannot be directly applied to LLMs due to their complexity
advanced capabilities. Second, as LLMs are increasingly deployed across diverse
industry applications, the role of XAI shifts from merely opening the "black
box" to actively enhancing the productivity and applicability of LLMs in
real-world settings. Meanwhile, unlike traditional machine learning models that
are passive recipients of XAI insights, the distinct abilities of LLMs can
reciprocally enhance XAI. Therefore, in this paper, we introduce Usable XAI in
the context of LLMs by analyzing (1) how XAI can benefit LLMs and AI systems,
and (2) how LLMs can contribute to the advancement of XAI. We introduce 10
strategies, introducing the key techniques for each and discussing their
associated challenges. We also provide case studies to demonstrate how to
obtain and leverage explanations. The code used in this paper can be found at:
https://github.com/JacksonWuxs/UsableXAI_LLM.
|
cs.LG
|
Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Era
|
2024-03-13T20:19:30Z
|
http://arxiv.org/pdf/2403.08943v1
|
http://arxiv.org/abs/2403.08943v1
|
Jianlin Chen
|
2024-03-13T20:19:30Z
|
cs.CL
|
Since the breakthrough of ChatGPT, large language models (LLMs) have garnered
significant attention in the research community. With the development of LLMs,
the question of text style transfer for conversational models has emerged as a
natural extension, where chatbots may possess their own styles or even
characters. However, standard evaluation metrics have not yet been established
for this new settings. This paper aims to address this issue by proposing the
LMStyle Benchmark, a novel evaluation framework applicable to chat-style text
style transfer (C-TST), that can measure the quality of style transfer for LLMs
in an automated and scalable manner. In addition to conventional style strength
metrics, LMStyle Benchmark further considers a novel aspect of metrics called
appropriateness, a high-level metrics take account of coherence, fluency and
other implicit factors without the aid of reference samples. Our experiments
demonstrate that the new evaluation methods introduced by LMStyle Benchmark
have a higher correlation with human judgments in terms of appropriateness.
Based on LMStyle Benchmark, we present a comprehensive list of evaluation
results for popular LLMs, including LLaMA, Alpaca, and Vicuna, reflecting their
stylistic properties, such as formality and sentiment strength, along with
their appropriateness.
|
cs.CL
|
LMStyle Benchmark: Evaluating Text Style Transfer for Chatbots
|
2024-03-13T18:57:30Z
|
http://arxiv.org/pdf/2403.10557v1
|
http://arxiv.org/abs/2403.10557v1
|
Kang Gu, Md Rafi Ur Rashid, Najrin Sultana, Shagufta Mehnaz
|
2024-03-13T18:57:30Z
|
cs.LG, cs.AI, cs.CL
|
With the rapid development of Large Language Models (LLMs), we have witnessed
intense competition among the major LLM products like ChatGPT, LLaMa, and
Gemini. However, various issues (e.g. privacy leakage and copyright violation)
of the training corpus still remain underexplored. For example, the Times sued
OpenAI and Microsoft for infringing on its copyrights by using millions of its
articles for training. From the perspective of LLM practitioners, handling such
unintended privacy violations can be challenging. Previous work addressed the
``unlearning" problem of LLMs using gradient information, while they mostly
introduced significant overheads like data preprocessing or lacked robustness.
In this paper, contrasting with the methods based on first-order information,
we revisit the unlearning problem via the perspective of second-order
information (Hessian). Our unlearning algorithms, which are inspired by classic
Newton update, are not only data-agnostic/model-agnostic but also proven to be
robust in terms of utility preservation or privacy guarantee. Through a
comprehensive evaluation with four NLP datasets as well as a case study on
real-world datasets, our methods consistently show superiority over the
first-order methods.
|
cs.LG
|
Second-Order Information Matters: Revisiting Machine Unlearning for Large Language Models
|
2024-03-13T18:47:00Z
|
http://arxiv.org/pdf/2403.08904v1
|
http://arxiv.org/abs/2403.08904v1
|
Tyler A. Chang, Katrin Tomanek, Jessica Hoffmann, Nithum Thain, Erin van Liemt, Kathleen Meier-Hellstern, Lucas Dixon
|
2024-03-13T18:47:00Z
|
cs.CL
|
We explore a strategy to handle controversial topics in LLM-based chatbots
based on Wikipedia's Neutral Point of View (NPOV) principle: acknowledge the
absence of a single true answer and surface multiple perspectives. We frame
this as retrieval augmented generation, where perspectives are retrieved from a
knowledge base and the LLM is tasked with generating a fluent and faithful
response from the given perspectives. As a starting point, we use a
deterministic retrieval system and then focus on common LLM failure modes that
arise during this approach to text generation, namely hallucination and
coverage errors. We propose and evaluate three methods to detect such errors
based on (1) word-overlap, (2) salience, and (3) LLM-based classifiers. Our
results demonstrate that LLM-based classifiers, even when trained only on
synthetic errors, achieve high error detection performance, with ROC AUC scores
of 95.3% for hallucination and 90.5% for coverage error detection on
unambiguous error cases. We show that when no training data is available, our
other methods still yield good results on hallucination (84.0%) and coverage
error (85.2%) detection.
|
cs.CL
|
Detecting Hallucination and Coverage Errors in Retrieval Augmented Generation for Controversial Topics
|
2024-03-13T18:20:24Z
|
http://arxiv.org/pdf/2403.08890v1
|
http://arxiv.org/abs/2403.08890v1
|
Claire Augusta Bergey, Simon DeDeo
|
2024-03-13T18:20:24Z
|
cs.CL, cs.IT, math.IT, q-bio.NC
|
Conversation demands attention. Speakers must call words to mind, listeners
must make sense of them, and both together must negotiate this flow of
information, all in fractions of a second. We used large language models to
study how this works in a large-scale dataset of English-language conversation,
the CANDOR corpus. We provide a new estimate of the information density of
unstructured conversation, of approximately 13 bits/second, and find
significant effects associated with the cognitive load of both retrieving, and
presenting, that information. We also reveal a role for backchannels -- the
brief yeahs, uh-huhs, and mhmms that listeners provide -- in regulating the
production of novelty: the lead-up to a backchannel is associated with
declining information rate, while speech downstream rebounds to previous rates.
Our results provide new insights into long-standing theories of how we respond
to fluctuating demands on cognitive resources, and how we negotiate those
demands in partnership with others.
|
cs.CL
|
From "um" to "yeah": Producing, predicting, and regulating information flow in human conversation
|
2024-03-13T18:16:21Z
|
http://arxiv.org/pdf/2403.09738v4
|
http://arxiv.org/abs/2403.09738v4
|
Se-eun Yoon, Zhankui He, Jessica Maria Echterhoff, Julian McAuley
|
2024-03-25T23:53:01Z
|
cs.CL, cs.AI, cs.IR
|
Synthetic users are cost-effective proxies for real users in the evaluation
of conversational recommender systems. Large language models show promise in
simulating human-like behavior, raising the question of their ability to
represent a diverse population of users. We introduce a new protocol to measure
the degree to which language models can accurately emulate human behavior in
conversational recommendation. This protocol is comprised of five tasks, each
designed to evaluate a key property that a synthetic user should exhibit:
choosing which items to talk about, expressing binary preferences, expressing
open-ended preferences, requesting recommendations, and giving feedback.
Through evaluation of baseline simulators, we demonstrate these tasks
effectively reveal deviations of language models from human behavior, and offer
insights on how to reduce the deviations with model selection and prompting
strategies.
|
cs.CL
|
Evaluating Large Language Models as Generative User Simulators for Conversational Recommendation
|
2024-03-13T18:00:00Z
|
http://arxiv.org/pdf/2403.08851v1
|
http://arxiv.org/abs/2403.08851v1
|
Siddharth Mishra-Sharma, Yiding Song, Jesse Thaler
|
2024-03-13T18:00:00Z
|
astro-ph.IM, cs.CL, cs.CV, cs.IR, cs.LG
|
We present PAPERCLIP (Proposal Abstracts Provide an Effective Representation
for Contrastive Language-Image Pre-training), a method which associates
astronomical observations imaged by telescopes with natural language using a
neural network model. The model is fine-tuned from a pre-trained Contrastive
Language-Image Pre-training (CLIP) model using successful observing proposal
abstracts and corresponding downstream observations, with the abstracts
optionally summarized via guided generation using large language models (LLMs).
Using observations from the Hubble Space Telescope (HST) as an example, we show
that the fine-tuned model embodies a meaningful joint representation between
observations and natural language through tests targeting image retrieval
(i.e., finding the most relevant observations using natural language queries)
and description retrieval (i.e., querying for astrophysical object classes and
use cases most relevant to a given observation). Our study demonstrates the
potential for using generalist foundation models rather than task-specific
models for interacting with astronomical data by leveraging text as an
interface.
|
astro-ph.IM
|
PAPERCLIP: Associating Astronomical Observations and Natural Language with Multi-Modal Models
|
2024-03-13T17:58:57Z
|
http://arxiv.org/pdf/2403.08763v3
|
http://arxiv.org/abs/2403.08763v3
|
Adam Ibrahim, Benjamin Thérien, Kshitij Gupta, Mats L. Richter, Quentin Anthony, Timothée Lesort, Eugene Belilovsky, Irina Rish
|
2024-03-26T17:58:48Z
|
cs.LG, cs.AI, cs.CL
|
Large language models (LLMs) are routinely pre-trained on billions of tokens,
only to start the process over again once new data becomes available. A much
more efficient solution is to continually pre-train these models, saving
significant compute compared to re-training. However, the distribution shift
induced by new data typically results in degraded performance on previous data
or poor adaptation to the new data. In this work, we show that a simple and
scalable combination of learning rate (LR) re-warming, LR re-decaying, and
replay of previous data is sufficient to match the performance of fully
re-training from scratch on all available data, as measured by the final loss
and the average score on several language model (LM) evaluation benchmarks.
Specifically, we show this for a weak but realistic distribution shift between
two commonly used LLM pre-training datasets (English$\rightarrow$English) and a
stronger distribution shift (English$\rightarrow$German) at the $405$M
parameter model scale with large dataset sizes (hundreds of billions of
tokens). Selecting the weak but realistic shift for larger-scale experiments,
we also find that our continual learning strategies match the re-training
baseline for a 10B parameter LLM. Our results demonstrate that LLMs can be
successfully updated via simple and scalable continual learning strategies,
matching the re-training baseline using only a fraction of the compute.
Finally, inspired by previous work, we propose alternatives to the cosine
learning rate schedule that help circumvent forgetting induced by LR re-warming
and that are not bound to a fixed token budget.
|
cs.LG
|
Simple and Scalable Strategies to Continually Pre-train Large Language Models
|
2024-03-13T17:53:47Z
|
http://arxiv.org/pdf/2403.08755v2
|
http://arxiv.org/abs/2403.08755v2
|
Feng Cheng, Ziyang Wang, Yi-Lin Sung, Yan-Bo Lin, Mohit Bansal, Gedas Bertasius
|
2024-04-22T19:17:49Z
|
cs.CV, cs.AI, cs.CL, cs.LG
|
We present a parameter-efficient method for continual video
question-answering (VidQA) learning. Our method, named DAM, uses the proposed
Dynamic Adapter Merging to (i) mitigate catastrophic forgetting, (ii) enable
efficient adaptation to continually arriving datasets, (iii) handle inputs from
unknown datasets during inference, and (iv) enable knowledge sharing across
similar dataset domains. Given a set of continually streaming VidQA datasets,
we sequentially train dataset-specific adapters for each dataset while freezing
the parameters of a large pretrained video-language backbone. During inference,
given a video-question sample from an unknown domain, our method first uses the
proposed non-parametric router function to compute a probability for each
adapter, reflecting how relevant that adapter is to the current video-question
input instance. Subsequently, the proposed dynamic adapter merging scheme
aggregates all the adapter weights into a new adapter instance tailored for
that particular test sample to compute the final VidQA prediction, mitigating
the impact of inaccurate router predictions and facilitating knowledge sharing
across domains. Our DAM model outperforms prior state-of-the-art continual
learning approaches by 9.1% while exhibiting 1.9% less forgetting on 6 VidQA
datasets spanning various domains. We further extend DAM to continual image
classification and image QA and outperform prior methods by a large margin. The
code is publicly available at: https://github.com/klauscc/DAM
|
cs.CV
|
DAM: Dynamic Adapter Merging for Continual Video QA Learning
|
2024-03-13T17:46:28Z
|
http://arxiv.org/pdf/2403.08743v1
|
http://arxiv.org/abs/2403.08743v1
|
Jingling Li, Zeyu Tang, Xiaoyu Liu, Peter Spirtes, Kun Zhang, Liu Leqi, Yang Liu
|
2024-03-13T17:46:28Z
|
cs.CL, cs.AI, cs.LG
|
Large language models (LLMs) can easily generate biased and discriminative
responses. As LLMs tap into consequential decision-making (e.g., hiring and
healthcare), it is of crucial importance to develop strategies to mitigate
these biases. This paper focuses on social bias, tackling the association
between demographic information and LLM outputs. We propose a causality-guided
debiasing framework that utilizes causal understandings of (1) the
data-generating process of the training corpus fed to LLMs, and (2) the
internal reasoning process of LLM inference, to guide the design of prompts for
debiasing LLM outputs through selection mechanisms. Our framework unifies
existing de-biasing prompting approaches such as inhibitive instructions and
in-context contrastive examples, and sheds light on new ways of debiasing by
encouraging bias-free reasoning. Our strong empirical performance on real-world
datasets demonstrates that our framework provides principled guidelines on
debiasing LLM outputs even with only the black-box access.
|
cs.CL
|
Steering LLMs Towards Unbiased Responses: A Causality-Guided Debiasing Framework
|
2024-03-13T17:42:32Z
|
http://arxiv.org/pdf/2403.08739v1
|
http://arxiv.org/abs/2403.08739v1
|
Carlo Nicolini, Jacopo Staiano, Bruno Lepri, Raffaele Marino
|
2024-03-13T17:42:32Z
|
cs.CL, cond-mat.dis-nn, cond-mat.stat-mech, cs.AI
|
A substantial gap persists in understanding the reasons behind the
exceptional performance of the Transformer architecture in NLP. A particularly
unexplored area involves the mechanistic description of how the distribution of
parameters evolves over time during training. In this work we suggest that
looking at the time evolution of the statistic distribution of model
parameters, and specifically at bifurcation effects, can help understanding the
model quality, potentially reducing training costs and evaluation efforts and
empirically showing the reasons behind the effectiveness of weights
sparsification.
|
cs.CL
|
The Garden of Forking Paths: Observing Dynamic Parameters Distribution in Large Language Models
|
2024-03-13T17:42:03Z
|
http://arxiv.org/pdf/2403.08738v1
|
http://arxiv.org/abs/2403.08738v1
|
Amit Meghanani, Thomas Hain
|
2024-03-13T17:42:03Z
|
cs.CL, cs.SD, eess.AS
|
Acoustic word embeddings (AWEs) are vector representations of spoken words.
An effective method for obtaining AWEs is the Correspondence Auto-Encoder
(CAE). In the past, the CAE method has been associated with traditional MFCC
features. Representations obtained from self-supervised learning (SSL)-based
speech models such as HuBERT, Wav2vec2, etc., are outperforming MFCC in many
downstream tasks. However, they have not been well studied in the context of
learning AWEs. This work explores the effectiveness of CAE with SSL-based
speech representations to obtain improved AWEs. Additionally, the capabilities
of SSL-based speech models are explored in cross-lingual scenarios for
obtaining AWEs. Experiments are conducted on five languages: Polish,
Portuguese, Spanish, French, and English. HuBERT-based CAE model achieves the
best results for word discrimination in all languages, despite Hu-BERT being
pre-trained on English only. Also, the HuBERT-based CAE model works well in
cross-lingual settings. It outperforms MFCC-based CAE models trained on the
target languages when trained on one source language and tested on target
languages.
|
cs.CL
|
Improving Acoustic Word Embeddings through Correspondence Training of Self-supervised Speech Representations
|
2024-03-13T17:38:05Z
|
http://arxiv.org/pdf/2403.08737v1
|
http://arxiv.org/abs/2403.08737v1
|
Sayar Ghosh Roy, Jiawei Han
|
2024-03-13T17:38:05Z
|
cs.IR, cs.CL
|
Existing Machine Learning approaches for local citation recommendation
directly map or translate a query, which is typically a claim or an entity
mention, to citation-worthy research papers. Within such a formulation, it is
challenging to pinpoint why one should cite a specific research paper for a
particular query, leading to limited recommendation interpretability. To
alleviate this, we introduce the evidence-grounded local citation
recommendation task, where the target latent space comprises evidence spans for
recommending specific papers. Using a distantly-supervised evidence retrieval
and multi-step re-ranking framework, our proposed system, ILCiteR, recommends
papers to cite for a query grounded on similar evidence spans extracted from
the existing research literature. Unlike past formulations that simply output
recommendations, ILCiteR retrieves ranked lists of evidence span and
recommended paper pairs. Secondly, previously proposed neural models for
citation recommendation require expensive training on massive labeled data,
ideally after every significant update to the pool of candidate papers. In
contrast, ILCiteR relies solely on distant supervision from a dynamic evidence
database and pre-trained Transformer-based Language Models without any model
training. We contribute a novel dataset for the evidence-grounded local
citation recommendation task and demonstrate the efficacy of our proposed
conditional neural rank-ensembling approach for re-ranking evidence spans.
|
cs.IR
|
ILCiteR: Evidence-grounded Interpretable Local Citation Recommendation
|
2024-03-13T17:29:45Z
|
http://arxiv.org/pdf/2403.08730v2
|
http://arxiv.org/abs/2403.08730v2
|
Renjie Pi, Tianyang Han, Wei Xiong, Jipeng Zhang, Runtao Liu, Rui Pan, Tong Zhang
|
2024-04-03T15:22:23Z
|
cs.CL, cs.CV
|
Multimodal Large Language Models (MLLMs) excel in generating responses based
on visual inputs. However, they often suffer from a bias towards generating
responses similar to their pretraining corpus, overshadowing the importance of
visual information. We treat this bias as a "preference" for pretraining
statistics, which hinders the model's grounding in visual input. To mitigate
this issue, we propose Bootstrapped Preference Optimization (BPO), which
conducts preference learning with datasets containing negative responses
bootstrapped from the model itself. Specifically, we propose the following two
strategies: 1) using distorted image inputs to the MLLM for eliciting responses
that contain signified pretraining bias; 2) leveraging text-based LLM to
explicitly inject erroneous but common elements into the original response.
Those undesirable responses are paired with original annotated responses from
the datasets to construct the preference dataset, which is subsequently
utilized to perform preference learning. Our approach effectively suppresses
pretrained LLM bias, enabling enhanced grounding in visual inputs. Extensive
experimentation demonstrates significant performance improvements across
multiple benchmarks, advancing the state-of-the-art in multimodal
conversational systems.
|
cs.CL
|
Strengthening Multimodal Large Language Model with Bootstrapped Preference Optimization
|
2024-03-13T17:17:48Z
|
http://arxiv.org/pdf/2403.08715v3
|
http://arxiv.org/abs/2403.08715v3
|
Ruiyi Wang, Haofei Yu, Wenxin Zhang, Zhengyang Qi, Maarten Sap, Graham Neubig, Yonatan Bisk, Hao Zhu
|
2024-04-25T20:23:41Z
|
cs.CL
|
Humans learn social skills through both imitation and social interaction.
This social learning process is largely understudied by existing research on
building language agents. Motivated by this gap, we propose an interactive
learning method, SOTOPIA-$\pi$, improving the social intelligence of language
agents. This method leverages behavior cloning and self-reinforcement training
on filtered social interaction data according to large language model (LLM)
ratings. We show that our training method allows a 7B LLM to reach the social
goal completion ability of an expert model (GPT-4-based agent), while improving
the safety of language agents and maintaining general QA ability on the MMLU
benchmark. We also find that this training paradigm uncovers some difficulties
in LLM-based evaluation of social intelligence: LLM-based evaluators
overestimate the abilities of the language agents trained specifically for
social interaction.
|
cs.CL
|
SOTOPIA-$π$: Interactive Learning of Socially Intelligent Language Agents
|
2024-03-13T16:57:57Z
|
http://arxiv.org/pdf/2403.08694v2
|
http://arxiv.org/abs/2403.08694v2
|
Shangding Gu, Alois Knoll, Ming Jin
|
2024-05-03T22:44:24Z
|
cs.CL
|
The development of Large Language Models (LLMs) often confronts challenges
stemming from the heavy reliance on human annotators in the reinforcement
learning with human feedback (RLHF) framework, or the frequent and costly
external queries tied to the self-instruct paradigm. In this work, we pivot to
Reinforcement Learning (RL) -- but with a twist. Diverging from the typical
RLHF, which refines LLMs following instruction data training, we use RL to
directly generate the foundational instruction dataset that alone suffices for
fine-tuning. Our method, TeaMs-RL, uses a suite of textual operations and
rules, prioritizing the diversification of training datasets. It facilitates
the generation of high-quality data without excessive reliance on external
advanced models, paving the way for a single fine-tuning step and negating the
need for subsequent RLHF stages. Our findings highlight key advantages of our
approach: reduced need for human involvement and fewer model queries (only
$5.73\%$ of WizardLM's total), along with enhanced capabilities of LLMs in
crafting and comprehending complex instructions compared to strong baselines,
and substantially improved model privacy protection.
|
cs.CL
|
TeaMs-RL: Teaching LLMs to Teach Themselves Better Instructions via Reinforcement Learning
|
2024-03-13T16:56:33Z
|
http://arxiv.org/pdf/2403.08693v1
|
http://arxiv.org/abs/2403.08693v1
|
Rik van Noord, Taja Kuzman, Peter Rupnik, Nikola Ljubešić, Miquel Esplà-Gomis, Gema Ramírez-Sánchez, Antonio Toral
|
2024-03-13T16:56:33Z
|
cs.CL
|
Large, curated, web-crawled corpora play a vital role in training language
models (LMs). They form the lion's share of the training data in virtually all
recent LMs, such as the well-known GPT, LLaMA and XLM-RoBERTa models. However,
despite this importance, relatively little attention has been given to the
quality of these corpora. In this paper, we compare four of the currently most
relevant large, web-crawled corpora (CC100, MaCoCu, mC4 and OSCAR) across
eleven lower-resourced European languages. Our approach is two-fold: first, we
perform an intrinsic evaluation by performing a human evaluation of the quality
of samples taken from different corpora; then, we assess the practical impact
of the qualitative differences by training specific LMs on each of the corpora
and evaluating their performance on downstream tasks. We find that there are
clear differences in quality of the corpora, with MaCoCu and OSCAR obtaining
the best results. However, during the extrinsic evaluation, we actually find
that the CC100 corpus achieves the highest scores. We conclude that, in our
experiments, the quality of the web-crawled corpora does not seem to play a
significant role when training LMs.
|
cs.CL
|
Do Language Models Care About Text Quality? Evaluating Web-Crawled Corpora Across 11 Languages
|
2024-03-13T16:44:39Z
|
http://arxiv.org/pdf/2403.08688v1
|
http://arxiv.org/abs/2403.08688v1
|
Ben Athiwaratkun, Shiqi Wang, Mingyue Shang, Yuchen Tian, Zijian Wang, Sujan Kumar Gonugondla, Sanjay Krishna Gouda, Rob Kwiatowski, Ramesh Nallapati, Bing Xiang
|
2024-03-13T16:44:39Z
|
cs.CL, cs.AI
|
Generative models, widely utilized in various applications, can often
struggle with prompts corresponding to partial tokens. This struggle stems from
tokenization, where partial tokens fall out of distribution during inference,
leading to incorrect or nonsensical outputs. This paper examines a technique to
alleviate the tokenization artifact on text completion in generative models,
maintaining performance even in regular non-subword cases. The method, termed
token alignment, involves backtracking to the last complete tokens and ensuring
the model's generation aligns with the prompt. This approach showcases marked
improvement across many partial token scenarios, including nuanced cases like
space-prefix and partial indentation, with only a minor time increase. The
technique and analysis detailed in this paper contribute to the continuous
advancement of generative models in handling partial inputs, bearing relevance
for applications like code completion and text autocompletion.
|
cs.CL
|
Token Alignment via Character Matching for Subword Completion
|
2024-03-13T16:17:09Z
|
http://arxiv.org/pdf/2403.08664v2
|
http://arxiv.org/abs/2403.08664v2
|
Erlend Frayling, Jake Lever, Graham McDonald
|
2024-03-14T15:57:59Z
|
cs.CL, cs.LG
|
The challenge of accessing historical patient data for clinical research,
while adhering to privacy regulations, is a significant obstacle in medical
science. An innovative approach to circumvent this issue involves utilising
synthetic medical records that mirror real patient data without compromising
individual privacy. The creation of these synthetic datasets, particularly
without using actual patient data to train Large Language Models (LLMs),
presents a novel solution as gaining access to sensitive patient information to
train models is also a challenge. This study assesses the capability of the
Llama 2 LLM to create synthetic medical records that accurately reflect real
patient information, employing zero-shot and few-shot prompting strategies for
comparison against fine-tuned methodologies that do require sensitive patient
data during training. We focus on generating synthetic narratives for the
History of Present Illness section, utilising data from the MIMIC-IV dataset
for comparison. In this work introduce a novel prompting technique that
leverages a chain-of-thought approach, enhancing the model's ability to
generate more accurate and contextually relevant medical narratives without
prior fine-tuning. Our findings suggest that this chain-of-thought prompted
approach allows the zero-shot model to achieve results on par with those of
fine-tuned models, based on Rouge metrics evaluation.
|
cs.CL
|
Zero-shot and Few-shot Generation Strategies for Artificial Clinical Records
|
2024-03-13T15:38:55Z
|
http://arxiv.org/pdf/2403.15430v1
|
http://arxiv.org/abs/2403.15430v1
|
Jesse Atuhurra, Seiveright Cargill Dujohn, Hidetaka Kamigaito, Hiroyuki Shindo, Taro Watanabe
|
2024-03-13T15:38:55Z
|
cs.CL
|
Natural language processing (NLP) practitioners are leveraging large language
models (LLM) to create structured datasets from semi-structured and
unstructured data sources such as patents, papers, and theses, without having
domain-specific knowledge. At the same time, ecological experts are searching
for a variety of means to preserve biodiversity. To contribute to these
efforts, we focused on endangered species and through in-context learning, we
distilled knowledge from GPT-4. In effect, we created datasets for both named
entity recognition (NER) and relation extraction (RE) via a two-stage process:
1) we generated synthetic data from GPT-4 of four classes of endangered
species, 2) humans verified the factual accuracy of the synthetic data,
resulting in gold data. Eventually, our novel dataset contains a total of 3.6K
sentences, evenly divided between 1.8K NER and 1.8K RE sentences. The
constructed dataset was then used to fine-tune both general BERT and
domain-specific BERT variants, completing the knowledge distillation process
from GPT-4 to BERT, because GPT-4 is resource intensive. Experiments show that
our knowledge transfer approach is effective at creating a NER model suitable
for detecting endangered species from texts.
|
cs.CL
|
Distilling Named Entity Recognition Models for Endangered Species from Large Language Models
|
2024-03-13T15:20:30Z
|
http://arxiv.org/pdf/2403.08607v1
|
http://arxiv.org/abs/2403.08607v1
|
Subash Neupane, Shaswata Mitra, Sudip Mittal, Noorbakhsh Amiri Golilarz, Shahram Rahimi, Amin Amirlatifi
|
2024-03-13T15:20:30Z
|
cs.CL, cs.AI
|
Large Language Models (LLMs) have shown impressive capabilities in generating
human-like responses. However, their lack of domain-specific knowledge limits
their applicability in healthcare settings, where contextual and comprehensive
responses are vital. To address this challenge and enable the generation of
patient-centric responses that are contextually relevant and comprehensive, we
propose MedInsight:a novel retrieval augmented framework that augments LLM
inputs (prompts) with relevant background information from multiple sources.
MedInsight extracts pertinent details from the patient's medical record or
consultation transcript. It then integrates information from authoritative
medical textbooks and curated web resources based on the patient's health
history and condition. By constructing an augmented context combining the
patient's record with relevant medical knowledge, MedInsight generates
enriched, patient-specific responses tailored for healthcare applications such
as diagnosis, treatment recommendations, or patient education. Experiments on
the MTSamples dataset validate MedInsight's effectiveness in generating
contextually appropriate medical responses. Quantitative evaluation using the
Ragas metric and TruLens for answer similarity and answer correctness
demonstrates the model's efficacy. Furthermore, human evaluation studies
involving Subject Matter Expert (SMEs) confirm MedInsight's utility, with
moderate inter-rater agreement on the relevance and correctness of the
generated responses.
|
cs.CL
|
MedInsight: A Multi-Source Context Augmentation Framework for Generating Patient-Centric Medical Responses using Large Language Models
|
2024-03-13T15:13:44Z
|
http://arxiv.org/pdf/2403.08604v2
|
http://arxiv.org/abs/2403.08604v2
|
Bowen Li, Wenhan Wu, Ziwei Tang, Lin Shi, John Yang, Jinyang Li, Shunyu Yao, Chen Qian, Binyuan Hui, Qicheng Zhang, Zhiyin Yu, He Du, Ping Yang, Dahua Lin, Chao Peng, Kai Chen
|
2024-03-15T13:23:34Z
|
cs.CL, cs.SE
|
Recent advancements in large language models (LLMs) have significantly
enhanced their coding capabilities. However, existing benchmarks predominantly
focused on simplified or isolated aspects of programming, such as single-file
code generation or repository issue debugging, falling short of measuring the
full spectrum of challenges raised by real-world programming activities. To
this end, we propose DevBench, a comprehensive benchmark that evaluates LLMs
across various stages of the software development lifecycle, including software
design, environment setup, implementation, acceptance testing, and unit
testing. DevBench features a wide range of programming languages and domains,
high-quality data collection, and carefully designed and verified metrics for
each task. Empirical studies show that current LLMs, including GPT-4-Turbo,
fail to solve the challenges presented within DevBench. Analyses reveal that
models struggle with understanding the complex structures in the repository,
managing the compilation process, and grasping advanced programming concepts.
Our findings offer actionable insights for the future development of LLMs
toward real-world programming applications. Our benchmark is available at
https://github.com/open-compass/DevBench
|
cs.CL
|
DevBench: A Comprehensive Benchmark for Software Development
|
2024-03-13T14:59:07Z
|
http://arxiv.org/pdf/2403.08593v2
|
http://arxiv.org/abs/2403.08593v2
|
Sitao Cheng, Ziyuan Zhuang, Yong Xu, Fangkai Yang, Chaoyun Zhang, Xiaoting Qin, Xiang Huang, Ling Chen, Qingwei Lin, Dongmei Zhang, Saravan Rajmohan, Qi Zhang
|
2024-07-03T15:23:59Z
|
cs.CL, cs.AI
|
Large Language Models (LLMs) have shown potential in reasoning over
structured environments, e.g., knowledge graph and table. Such tasks typically
require multi-hop reasoning, i.e., match natural language utterance with
instances in the environment. Previous methods leverage LLMs to incrementally
build a reasoning path, where the LLMs either invoke tools or pick up schemas
by step-by-step interacting with the environment. We propose
Reasoning-Path-Editing (Readi), a novel framework where LLMs can efficiently
and faithfully reason over structured environments. In Readi, LLMs initially
generate a reasoning path given a query, and edit the path only when necessary.
We instantiate the path on structured environments and provide feedback to edit
the path if anything goes wrong. Experimental results on three KGQA and two
TableQA datasets show the effectiveness of Readi, significantly surpassing
previous LLM-based methods (by 9.1% Hit@1 on WebQSP, 12.4% on MQA-3H and 9.5%
on WTQ), comparable with state-of-the-art fine-tuned methods (67% on CWQ and
74.7% on WebQSP) and substantially boosting the vanilla LLMs (by 14.9% on CWQ).
Our code will be available on https://aka.ms/readi.
|
cs.CL
|
Call Me When Necessary: LLMs can Efficiently and Faithfully Reason over Structured Environments
|
2024-03-13T14:19:08Z
|
http://arxiv.org/pdf/2403.08564v1
|
http://arxiv.org/abs/2403.08564v1
|
Sara Sterlie, Nina Weng, Aasa Feragen
|
2024-03-13T14:19:08Z
|
cs.CL, cs.AI, cs.HC
|
Within recent years, generative AI, such as large language models, has
undergone rapid development. As these models become increasingly available to
the public, concerns arise about perpetuating and amplifying harmful biases in
applications. Gender stereotypes can be harmful and limiting for the
individuals they target, whether they consist of misrepresentation or
discrimination. Recognizing gender bias as a pervasive societal construct, this
paper studies how to uncover and quantify the presence of gender biases in
generative language models. In particular, we derive generative AI analogues of
three well-known non-discrimination criteria from classification, namely
independence, separation and sufficiency. To demonstrate these criteria in
action, we design prompts for each of the criteria with a focus on occupational
gender stereotype, specifically utilizing the medical test to introduce the
ground truth in the generative AI context. Our results address the presence of
occupational gender bias within such conversational language models.
|
cs.CL
|
Non-discrimination Criteria for Generative Language Models
|
2024-03-13T13:54:00Z
|
http://arxiv.org/pdf/2403.08540v2
|
http://arxiv.org/abs/2403.08540v2
|
Samir Yitzhak Gadre, Georgios Smyrnis, Vaishaal Shankar, Suchin Gururangan, Mitchell Wortsman, Rulin Shao, Jean Mercat, Alex Fang, Jeffrey Li, Sedrick Keh, Rui Xin, Marianna Nezhurina, Igor Vasiljevic, Jenia Jitsev, Luca Soldaini, Alexandros G. Dimakis, Gabriel Ilharco, Pang Wei Koh, Shuran Song, Thomas Kollar, Yair Carmon, Achal Dave, Reinhard Heckel, Niklas Muennighoff, Ludwig Schmidt
|
2024-06-14T20:21:05Z
|
cs.CL, cs.LG
|
Scaling laws are useful guides for derisking expensive training runs, as they
predict performance of large models using cheaper, small-scale experiments.
However, there remain gaps between current scaling studies and how language
models are ultimately trained and evaluated. For instance, scaling is usually
studied in the compute-optimal training regime (i.e., "Chinchilla optimal"
regime). In contrast, models are often over-trained to reduce inference costs.
Moreover, scaling laws mostly predict loss on next-token prediction, but models
are usually compared on downstream task performance. To address both
shortcomings, we create a testbed of 104 models with 0.011B to 6.9B parameters
trained with various numbers of tokens on three data distributions. First, we
fit scaling laws that extrapolate in both the amount of over-training and the
number of model parameters. This enables us to predict the validation loss of a
1.4B parameter, 900B token run (i.e., 32$\times$ over-trained) and a 6.9B
parameter, 138B token run (i.e., a compute-optimal run)$\unicode{x2014}$each
from experiments that take 300$\times$ less compute. Second, we relate the
perplexity of a language model to its downstream task performance by proposing
a power law. We use this law to predict top-1 error averaged over downstream
tasks for the two aforementioned models, using experiments that take 20$\times$
less compute. Our experiments are available at
https://github.com/mlfoundations/scaling.
|
cs.CL
|
Language models scale reliably with over-training and on downstream tasks
|
2024-03-13T13:04:58Z
|
http://arxiv.org/pdf/2403.08495v2
|
http://arxiv.org/abs/2403.08495v2
|
Yusheng Liao, Yutong Meng, Yuhao Wang, Hongcheng Liu, Yanfeng Wang, Yu Wang
|
2024-03-14T08:05:08Z
|
cs.CL
|
Large Language Models (LLMs) have demonstrated remarkable proficiency in
human interactions, yet their application within the medical field remains
insufficiently explored. Previous works mainly focus on the performance of
medical knowledge with examinations, which is far from the realistic scenarios,
falling short in assessing the abilities of LLMs on clinical tasks. In the
quest to enhance the application of Large Language Models (LLMs) in healthcare,
this paper introduces the Automated Interactive Evaluation (AIE) framework and
the State-Aware Patient Simulator (SAPS), targeting the gap between traditional
LLM evaluations and the nuanced demands of clinical practice. Unlike prior
methods that rely on static medical knowledge assessments, AIE and SAPS provide
a dynamic, realistic platform for assessing LLMs through multi-turn
doctor-patient simulations. This approach offers a closer approximation to real
clinical scenarios and allows for a detailed analysis of LLM behaviors in
response to complex patient interactions. Our extensive experimental validation
demonstrates the effectiveness of the AIE framework, with outcomes that align
well with human evaluations, underscoring its potential to revolutionize
medical LLM testing for improved healthcare delivery.
|
cs.CL
|
Automatic Interactive Evaluation for Large Language Models with State Aware Patient Simulator
|
2024-03-13T12:55:43Z
|
http://arxiv.org/pdf/2403.08492v2
|
http://arxiv.org/abs/2403.08492v2
|
Ming Dong, Yujing Chen, Miao Zhang, Hao Sun, Tingting He
|
2024-06-07T08:41:58Z
|
cs.CL
|
Chinese Spell Checking (CSC) is a widely used technology, which plays a vital
role in speech to text (STT) and optical character recognition (OCR). Most of
the existing CSC approaches relying on BERT architecture achieve excellent
performance. However, limited by the scale of the foundation model, BERT-based
method does not work well in few-shot scenarios, showing certain limitations in
practical applications. In this paper, we explore using an in-context learning
method named RS-LLM (Rich Semantic based LLMs) to introduce large language
models (LLMs) as the foundation model. Besides, we study the impact of
introducing various Chinese rich semantic information in our framework. We
found that by introducing a small number of specific Chinese rich semantic
structures, LLMs achieve better performance than the BERT-based model on
few-shot CSC task. Furthermore, we conduct experiments on multiple datasets,
and the experimental results verified the superiority of our proposed
framework.
|
cs.CL
|
Rich Semantic Knowledge Enhanced Large Language Models for Few-shot Chinese Spell Checking
|
2024-03-13T12:50:23Z
|
http://arxiv.org/pdf/2403.08484v1
|
http://arxiv.org/abs/2403.08484v1
|
Ming Dong, Kang Xue, Bolong Zheng, Tingting He
|
2024-03-13T12:50:23Z
|
cs.CL
|
In view of the huge number of parameters of Large language models (LLMs) ,
tuning all parameters is very costly, and accordingly fine-tuning specific
parameters is more sensible. Most of parameter efficient fine-tuning (PEFT)
concentrate on parameter selection strategies, such as additive method,
selective method and reparametrization-based method. However, there are few
methods that consider the impact of data samples on parameter selecting, such
as Fish Mask based method. Fish Mask randomly choose a part of data samples and
treat them equally during parameter selection, which is unable to dynamically
select optimal parameters for inconstant data distributions. In this work, we
adopt a data-oriented perspective, then proposing an IRD ($\mathrm{\underline
I}$terative sample-parameter $\mathrm{\underline R}$ange $\mathrm{\underline
D}$ecreasing) algorithm to search the best setting of sample-parameter pair for
FISH Mask. In each iteration, by searching the set of samples and parameters
with larger Fish information, IRD can find better sample-parameter pair in most
scale. We demonstrate the effectiveness and rationality of proposed strategy by
conducting experiments on GLUE benchmark. Experimental results show our
strategy optimizes the parameter selection and achieves preferable performance.
|
cs.CL
|
Data-oriented Dynamic Fine-tuning Parameter Selection Strategy for FISH Mask based Efficient Fine-tuning
|
2024-03-13T12:25:47Z
|
http://arxiv.org/pdf/2403.08462v1
|
http://arxiv.org/abs/2403.08462v1
|
Andrea Nini, Oren Halvani, Lukas Graner, Valerio Gherardi, Shunichi Ishihara
|
2024-03-13T12:25:47Z
|
cs.CL, cs.LG
|
Authorship Verification (AV) is the process of analyzing a set of documents
to determine whether they were written by a specific author. This problem often
arises in forensic scenarios, e.g., in cases where the documents in question
constitute evidence for a crime. Existing state-of-the-art AV methods use
computational solutions that are not supported by a plausible scientific
explanation for their functioning and that are often difficult for analysts to
interpret. To address this, we propose a method relying on calculating a
quantity we call $\lambda_G$ (LambdaG): the ratio between the likelihood of a
document given a model of the Grammar for the candidate author and the
likelihood of the same document given a model of the Grammar for a reference
population. These Grammar Models are estimated using $n$-gram language models
that are trained solely on grammatical features. Despite not needing large
amounts of data for training, LambdaG still outperforms other established AV
methods with higher computational complexity, including a fine-tuned Siamese
Transformer network. Our empirical evaluation based on four baseline methods
applied to twelve datasets shows that LambdaG leads to better results in terms
of both accuracy and AUC in eleven cases and in all twelve cases if considering
only topic-agnostic methods. The algorithm is also highly robust to important
variations in the genre of the reference population in many cross-genre
comparisons. In addition to these properties, we demonstrate how LambdaG is
easier to interpret than the current state-of-the-art. We argue that the
advantage of LambdaG over other methods is due to fact that it is compatible
with Cognitive Linguistic theories of language processing.
|
cs.CL
|
Authorship Verification based on the Likelihood Ratio of Grammar Models
|
2024-03-13T11:16:43Z
|
http://arxiv.org/pdf/2403.08424v1
|
http://arxiv.org/abs/2403.08424v1
|
Zeguan Xiao, Yan Yang, Guanhua Chen, Yun Chen
|
2024-03-13T11:16:43Z
|
cs.CR, cs.AI, cs.CL
|
Large language models (LLMs) have achieved significant advances in recent
days. Extensive efforts have been made before the public release of LLMs to
align their behaviors with human values. The primary goal of alignment is to
ensure their helpfulness, honesty and harmlessness. However, even meticulously
aligned LLMs remain vulnerable to malicious manipulations such as jailbreaking,
leading to unintended behaviors. The jailbreak is to intentionally develop a
malicious prompt that escapes from the LLM security restrictions to produce
uncensored detrimental contents. Previous works explore different jailbreak
methods for red teaming LLMs, yet they encounter challenges regarding to
effectiveness and scalability. In this work, we propose Tastle, a novel
black-box jailbreak framework for automated red teaming of LLMs. We designed
malicious content concealing and memory reframing with an iterative
optimization algorithm to jailbreak LLMs, motivated by the research about the
distractibility and over-confidence phenomenon of LLMs. Extensive experiments
of jailbreaking both open-source and proprietary LLMs demonstrate the
superiority of our framework in terms of effectiveness, scalability and
transferability. We also evaluate the effectiveness of existing jailbreak
defense methods against our attack and highlight the crucial need to develop
more effective and practical defense strategies.
|
cs.CR
|
Tastle: Distract Large Language Models for Automatic Jailbreak Attack
|
2024-03-13T10:10:07Z
|
http://arxiv.org/pdf/2403.08391v2
|
http://arxiv.org/abs/2403.08391v2
|
JooYoung Lee, Emily Booth, Hany Farid, Marian-Andrei Rizoiu
|
2024-05-26T06:15:33Z
|
cs.CL
|
What if misinformation is not an information problem at all? To understand
the role of news publishers in potentially unintentionally propagating
misinformation, we examine how far-right and fringe online groups share and
leverage established legacy news media articles to advance their narratives.
Our findings suggest that online fringe ideologies spread through the use of
content that is consensus-based and "factually correct". We found that
Australian news publishers with both moderate and far-right political leanings
contain comparable levels of information completeness and quality; and
furthermore, that far-right Twitter users often share from moderate sources.
However, a stark difference emerges when we consider two additional factors: 1)
the narrow topic selection of articles by far-right users, suggesting that they
cherry pick only news articles that engage with their preexisting worldviews
and specific topics of concern, and 2) the difference between moderate and
far-right publishers when we examine the writing style of their articles.
Furthermore, we can identify users prone to sharing misinformation based on
their communication style. These findings have important implications for
countering online misinformation, as they highlight the powerful role that
personal biases towards specific topics and publishers' writing styles have in
amplifying fringe ideologies online.
|
cs.CL
|
Misinformation is not about Bad Facts: An Analysis of the Production and Consumption of Fringe Content
|
2024-03-13T09:48:13Z
|
http://arxiv.org/pdf/2403.09734v2
|
http://arxiv.org/abs/2403.09734v2
|
Gustaw Opiełka, Hannes Rosenbusch, Veerle Vijverberg, Claire E. Stevenson
|
2024-05-13T11:20:23Z
|
cs.CL, cs.AI
|
The Abstraction Reasoning Corpus (ARC) is a visual analogical reasoning test
designed for humans and machines (Chollet, 2019). We compared human and large
language model (LLM) performance on a new child-friendly set of ARC items.
Results show that both children and adults outperform most LLMs on these tasks.
Error analysis revealed a similar "fallback" solution strategy in LLMs and
young children, where part of the analogy is simply copied. In addition, we
found two other error types, one based on seemingly grasping key concepts
(e.g., Inside-Outside) and the other based on simple combinations of analogy
input matrices. On the whole, "concept" errors were more common in humans, and
"matrix" errors were more common in LLMs. This study sheds new light on LLM
reasoning ability and the extent to which we can use error analyses and
comparisons with human development to understand how LLMs solve visual
analogies.
|
cs.CL
|
Do Large Language Models Solve ARC Visual Analogies Like People Do?
|
2024-03-13T09:42:46Z
|
http://arxiv.org/pdf/2403.08377v1
|
http://arxiv.org/abs/2403.08377v1
|
Fangqi Zhu, Yongqi Zhang, Lei Chen, Bing Qin, Ruifeng Xu
|
2024-03-13T09:42:46Z
|
cs.CL
|
Adverse drug-drug interactions~(DDIs) can compromise the effectiveness of
concurrent drug administration, posing a significant challenge in healthcare.
As the development of new drugs continues, the potential for unknown adverse
effects resulting from DDIs becomes a growing concern. Traditional
computational methods for DDI prediction may fail to capture interactions for
new drugs due to the lack of knowledge. In this paper, we introduce a new
problem setup as zero-shot DDI prediction that deals with the case of new
drugs. Leveraging textual information from online databases like DrugBank and
PubChem, we propose an innovative approach TextDDI with a language model-based
DDI predictor and a reinforcement learning~(RL)-based information selector,
enabling the selection of concise and pertinent text for accurate DDI
prediction on new drugs. Empirical results show the benefits of the proposed
approach on several settings including zero-shot and few-shot DDI prediction,
and the selected texts are semantically relevant. Our code and data are
available at \url{https://github.com/zhufq00/DDIs-Prediction}.
|
cs.CL
|
Learning to Describe for Predicting Zero-shot Drug-Drug Interactions
|
2024-03-13T09:38:39Z
|
http://arxiv.org/pdf/2403.08375v1
|
http://arxiv.org/abs/2403.08375v1
|
Ran Zmigrod, Salwa Alamir, Xiaomo Liu
|
2024-03-13T09:38:39Z
|
cs.DB, cs.AI, cs.CL, cs.SE
|
Migrations of systems from on-site premises to the cloud has been a
fundamental endeavor by many industrial institutions. A crucial component of
such cloud migrations is the transition of databases to be hosted online. In
this work, we consider the difficulties of this migration for SQL databases.
While SQL is one of the prominent methods for storing database procedures,
there are a plethora of different SQL dialects (e.g., MySQL, Postgres, etc.)
which can complicate migrations when the on-premise SQL dialect differs to the
dialect hosted on the cloud. Tools exist by common cloud provides such as AWS
and Azure to aid in translating between dialects in order to mitigate the
majority of the difficulties. However, these tools do not successfully
translate $100\%$ of the code. Consequently, software engineers must manually
convert the remainder of the untranslated database. For large organizations,
this task quickly becomes intractable and so more innovative solutions are
required. We consider this challenge a novel yet vital industrial research
problem for any large corporation that is considering cloud migrations.
Furthermore, we introduce potential avenues of research to tackle this
challenge that have yielded promising preliminary results.
|
cs.DB
|
Translating between SQL Dialects for Cloud Migration
|
2024-03-13T09:31:50Z
|
http://arxiv.org/pdf/2403.08370v1
|
http://arxiv.org/abs/2403.08370v1
|
H S V N S Kowndinya Renduchintala, Sumit Bhatia, Ganesh Ramakrishnan
|
2024-03-13T09:31:50Z
|
cs.CL, cs.AI, cs.LG
|
Instruction Tuning involves finetuning a language model on a collection of
instruction-formatted datasets in order to enhance the generalizability of the
model to unseen tasks. Studies have shown the importance of balancing different
task proportions during finetuning, but finding the right balance remains
challenging. Unfortunately, there's currently no systematic method beyond
manual tuning or relying on practitioners' intuition. In this paper, we
introduce SMART (Submodular data Mixture strAtegy for instRuction Tuning) - a
novel data mixture strategy which makes use of a submodular function to assign
importance scores to tasks which are then used to determine the mixture
weights. Given a fine-tuning budget, SMART redistributes the budget among tasks
and selects non-redundant samples from each task. Experimental results
demonstrate that SMART significantly outperforms traditional methods such as
examples proportional mixing and equal mixing. Furthermore, SMART facilitates
the creation of data mixtures based on a few representative subsets of tasks
alone and through task pruning analysis, we reveal that in a limited budget
setting, allocating budget among a subset of representative tasks yields
superior performance compared to distributing the budget among all tasks. The
code for reproducing our results is open-sourced at
https://github.com/kowndinya-renduchintala/SMART.
|
cs.CL
|
SMART: Submodular Data Mixture Strategy for Instruction Tuning
|
2024-03-13T09:18:46Z
|
http://arxiv.org/pdf/2403.08358v1
|
http://arxiv.org/abs/2403.08358v1
|
Rares Dolga, Ran Zmigrod, Rui Silva, Salwa Alamir, Sameena Shah
|
2024-03-13T09:18:46Z
|
cs.SE, cs.CL
|
Log analysis and monitoring are essential aspects in software maintenance and
identifying defects. In particular, the temporal nature and vast size of log
data leads to an interesting and important research question: How can logs be
summarised and monitored over time? While this has been a fundamental topic of
research in the software engineering community, work has typically focused on
heuristic-, syntax-, or static-based methods. In this work, we suggest an
online semantic-based clustering approach to error logs that dynamically
updates the log clusters to enable monitoring code error life-cycles. We also
introduce a novel metric to evaluate the performance of temporal log clusters.
We test our system and evaluation metric with an industrial dataset and find
that our solution outperforms similar systems. We hope that our work encourages
further temporal exploration in defect datasets.
|
cs.SE
|
Log Summarisation for Defect Evolution Analysis
|
2024-03-13T08:50:15Z
|
http://arxiv.org/pdf/2403.08345v1
|
http://arxiv.org/abs/2403.08345v1
|
Vamsi Krishna Kommineni, Birgitta König-Ries, Sheeba Samuel
|
2024-03-13T08:50:15Z
|
cs.CL
|
The conventional process of building Ontologies and Knowledge Graphs (KGs)
heavily relies on human domain experts to define entities and relationship
types, establish hierarchies, maintain relevance to the domain, fill the ABox
(or populate with instances), and ensure data quality (including amongst others
accuracy and completeness). On the other hand, Large Language Models (LLMs)
have recently gained popularity for their ability to understand and generate
human-like natural language, offering promising ways to automate aspects of
this process. This work explores the (semi-)automatic construction of KGs
facilitated by open-source LLMs. Our pipeline involves formulating competency
questions (CQs), developing an ontology (TBox) based on these CQs, constructing
KGs using the developed ontology, and evaluating the resultant KG with minimal
to no involvement of human experts. We showcase the feasibility of our
semi-automated pipeline by creating a KG on deep learning methodologies by
exploiting scholarly publications. To evaluate the answers generated via
Retrieval-Augmented-Generation (RAG) as well as the KG concepts automatically
extracted using LLMs, we design a judge LLM, which rates the generated content
based on ground truth. Our findings suggest that employing LLMs could
potentially reduce the human effort involved in the construction of KGs,
although a human-in-the-loop approach is recommended to evaluate automatically
generated KGs.
|
cs.CL
|
From human experts to machines: An LLM supported approach to ontology and knowledge graph construction
|
2024-03-13T08:34:53Z
|
http://arxiv.org/pdf/2403.08332v1
|
http://arxiv.org/abs/2403.08332v1
|
Heejin Do, Yunsu Kim, Gary Geunbae Lee
|
2024-03-13T08:34:53Z
|
cs.CL, cs.AI
|
Recently, encoder-only pre-trained models such as BERT have been successfully
applied in automated essay scoring (AES) to predict a single overall score.
However, studies have yet to explore these models in multi-trait AES, possibly
due to the inefficiency of replicating BERT-based models for each trait.
Breaking away from the existing sole use of encoder, we propose an
autoregressive prediction of multi-trait scores (ArTS), incorporating a
decoding process by leveraging the pre-trained T5. Unlike prior regression or
classification methods, we redefine AES as a score-generation task, allowing a
single model to predict multiple scores. During decoding, the subsequent trait
prediction can benefit by conditioning on the preceding trait scores.
Experimental results proved the efficacy of ArTS, showing over 5% average
improvements in both prompts and traits.
|
cs.CL
|
Autoregressive Score Generation for Multi-trait Essay Scoring
|
2024-03-13T08:02:23Z
|
http://arxiv.org/pdf/2403.08319v2
|
http://arxiv.org/abs/2403.08319v2
|
Rongwu Xu, Zehan Qi, Zhijiang Guo, Cunxiang Wang, Hongru Wang, Yue Zhang, Wei Xu
|
2024-06-22T08:31:40Z
|
cs.CL, cs.AI, cs.IR, cs.LG
|
This survey provides an in-depth analysis of knowledge conflicts for large
language models (LLMs), highlighting the complex challenges they encounter when
blending contextual and parametric knowledge. Our focus is on three categories
of knowledge conflicts: context-memory, inter-context, and intra-memory
conflict. These conflicts can significantly impact the trustworthiness and
performance of LLMs, especially in real-world applications where noise and
misinformation are common. By categorizing these conflicts, exploring the
causes, examining the behaviors of LLMs under such conflicts, and reviewing
available solutions, this survey aims to shed light on strategies for improving
the robustness of LLMs, thereby serving as a valuable resource for advancing
research in this evolving area.
|
cs.CL
|
Knowledge Conflicts for LLMs: A Survey
|
2024-03-13T07:52:31Z
|
http://arxiv.org/pdf/2403.09733v1
|
http://arxiv.org/abs/2403.09733v1
|
Haomin Wen, Zhenjie Wei, Yan Lin, Jiyuan Wang, Yuxuan Liang, Huaiyu Wan
|
2024-03-13T07:52:31Z
|
cs.CL, cs.AI
|
The rapid development of Large Language Models (LLMs) has facilitated a
variety of applications from different domains. In this technical report, we
explore the integration of LLMs and the popular academic writing tool,
Overleaf, to enhance the efficiency and quality of academic writing. To achieve
the above goal, there are three challenges: i) including seamless interaction
between Overleaf and LLMs, ii) establishing reliable communication with the LLM
provider, and iii) ensuring user privacy. To address these challenges, we
present OverleafCopilot, the first-ever tool (i.e., a browser extension) that
seamlessly integrates LLMs and Overleaf, enabling researchers to leverage the
power of LLMs while writing papers. Specifically, we first propose an effective
framework to bridge LLMs and Overleaf. Then, we developed PromptGenius, a
website for researchers to easily find and share high-quality up-to-date
prompts. Thirdly, we propose an agent command system to help researchers
quickly build their customizable agents. OverleafCopilot
(https://chromewebstore.google.com/detail/overleaf-copilot/eoadabdpninlhkkbhngoddfjianhlghb
) has been on the Chrome Extension Store, which now serves thousands of
researchers. Additionally, the code of PromptGenius is released at
https://github.com/wenhaomin/ChatGPT-PromptGenius. We believe our work has the
potential to revolutionize academic writing practices, empowering researchers
to produce higher-quality papers in less time.
|
cs.CL
|
OverleafCopilot: Empowering Academic Writing in Overleaf with Large Language Models
|
2024-03-13T07:49:50Z
|
http://arxiv.org/pdf/2403.08314v1
|
http://arxiv.org/abs/2403.08314v1
|
Sweta Agrawal, Amin Farajian, Patrick Fernandes, Ricardo Rei, André F. T. Martins
|
2024-03-13T07:49:50Z
|
cs.CL
|
Despite the recent success of automatic metrics for assessing translation
quality, their application in evaluating the quality of machine-translated
chats has been limited. Unlike more structured texts like news, chat
conversations are often unstructured, short, and heavily reliant on contextual
information. This poses questions about the reliability of existing
sentence-level metrics in this domain as well as the role of context in
assessing the translation quality. Motivated by this, we conduct a
meta-evaluation of existing sentence-level automatic metrics, primarily
designed for structured domains such as news, to assess the quality of
machine-translated chats. We find that reference-free metrics lag behind
reference-based ones, especially when evaluating translation quality in
out-of-English settings. We then investigate how incorporating conversational
contextual information in these metrics affects their performance. Our findings
show that augmenting neural learned metrics with contextual information helps
improve correlation with human judgments in the reference-free scenario and
when evaluating translations in out-of-English settings. Finally, we propose a
new evaluation metric, Context-MQM, that utilizes bilingual context with a
large language model (LLM) and further validate that adding context helps even
for LLM-based evaluation metrics.
|
cs.CL
|
Is Context Helpful for Chat Translation Evaluation?
|
2024-03-13T07:44:14Z
|
http://arxiv.org/pdf/2403.08312v1
|
http://arxiv.org/abs/2403.08312v1
|
Jia-Nan Li, Quan Tu, Cunli Mao, Zhengtao Yu, Ji-Rong Wen, Rui Yan
|
2024-03-13T07:44:14Z
|
cs.CL, cs.AI
|
Standard Large Language Models (LLMs) struggle with handling dialogues with
long contexts due to efficiency and consistency issues. According to our
observation, dialogue contexts are highly structured, and the special token of
\textit{End-of-Utterance} (EoU) in dialogues has the potential to aggregate
information. We refer to the EoU tokens as ``conversational attention sinks''
(conv-attn sinks). Accordingly, we introduce StreamingDialogue, which
compresses long dialogue history into conv-attn sinks with minimal losses, and
thus reduces computational complexity quadratically with the number of sinks
(i.e., the number of utterances). Current LLMs already demonstrate the ability
to handle long context window, e.g., a window size of 200k or more. To this
end, by compressing utterances into EoUs, our method has the potential to
handle more than 200k of utterances, resulting in a prolonged dialogue
learning. In order to minimize information losses from reconstruction after
compression, we design two learning strategies of short-memory reconstruction
(SMR) and long-memory reactivation (LMR). Our method outperforms strong
baselines in dialogue tasks and achieves a 4 $\times$ speedup while reducing
memory usage by 18 $\times$ compared to dense attention recomputation.
|
cs.CL
|
StreamingDialogue: Prolonged Dialogue Learning via Long Context Compression with Minimal Losses
|
2024-03-13T07:31:20Z
|
http://arxiv.org/pdf/2403.08305v1
|
http://arxiv.org/abs/2403.08305v1
|
Mingyue Cheng, Hao Zhang, Jiqian Yang, Qi Liu, Li Li, Xin Huang, Liwei Song, Zhi Li, Zhenya Huang, Enhong Chen
|
2024-03-13T07:31:20Z
|
cs.CL
|
Large language model evaluation plays a pivotal role in the enhancement of
its capacity. Previously, numerous methods for evaluating large language models
have been proposed in this area. Despite their effectiveness, these existing
works mainly focus on assessing objective questions, overlooking the capability
to evaluate subjective questions which is extremely common for large language
models. Additionally, these methods predominantly utilize centralized datasets
for evaluation, with question banks concentrated within the evaluation
platforms themselves. Moreover, the evaluation processes employed by these
platforms often overlook personalized factors, neglecting to consider the
individual characteristics of both the evaluators and the models being
evaluated. To address these limitations, we propose a novel anonymous
crowd-sourcing evaluation platform, BingJian, for large language models that
employs a competitive scoring mechanism where users participate in ranking
models based on their performance. This platform stands out not only for its
support of centralized evaluations to assess the general capabilities of models
but also for offering an open evaluation gateway. Through this gateway, users
have the opportunity to submit their questions, testing the models on a
personalized and potentially broader range of capabilities. Furthermore, our
platform introduces personalized evaluation scenarios, leveraging various forms
of human-computer interaction to assess large language models in a manner that
accounts for individual user preferences and contexts. The demonstration of
BingJian can be accessed at https://github.com/Mingyue-Cheng/Bingjian.
|
cs.CL
|
Towards Personalized Evaluation of Large Language Models with An Anonymous Crowd-Sourcing Platform
|
2024-03-13T06:59:16Z
|
http://arxiv.org/pdf/2403.08295v4
|
http://arxiv.org/abs/2403.08295v4
|
Gemma Team, Thomas Mesnard, Cassidy Hardin, Robert Dadashi, Surya Bhupatiraju, Shreya Pathak, Laurent Sifre, Morgane Rivière, Mihir Sanjay Kale, Juliette Love, Pouya Tafti, Léonard Hussenot, Pier Giuseppe Sessa, Aakanksha Chowdhery, Adam Roberts, Aditya Barua, Alex Botev, Alex Castro-Ros, Ambrose Slone, Amélie Héliou, Andrea Tacchetti, Anna Bulanova, Antonia Paterson, Beth Tsai, Bobak Shahriari, Charline Le Lan, Christopher A. Choquette-Choo, Clément Crepy, Daniel Cer, Daphne Ippolito, David Reid, Elena Buchatskaya, Eric Ni, Eric Noland, Geng Yan, George Tucker, George-Christian Muraru, Grigory Rozhdestvenskiy, Henryk Michalewski, Ian Tenney, Ivan Grishchenko, Jacob Austin, James Keeling, Jane Labanowski, Jean-Baptiste Lespiau, Jeff Stanway, Jenny Brennan, Jeremy Chen, Johan Ferret, Justin Chiu, Justin Mao-Jones, Katherine Lee, Kathy Yu, Katie Millican, Lars Lowe Sjoesund, Lisa Lee, Lucas Dixon, Machel Reid, Maciej Mikuła, Mateo Wirth, Michael Sharman, Nikolai Chinaev, Nithum Thain, Olivier Bachem, Oscar Chang, Oscar Wahltinez, Paige Bailey, Paul Michel, Petko Yotov, Rahma Chaabouni, Ramona Comanescu, Reena Jana, Rohan Anil, Ross McIlroy, Ruibo Liu, Ryan Mullins, Samuel L Smith, Sebastian Borgeaud, Sertan Girgin, Sholto Douglas, Shree Pandya, Siamak Shakeri, Soham De, Ted Klimenko, Tom Hennigan, Vlad Feinberg, Wojciech Stokowiec, Yu-hui Chen, Zafarali Ahmed, Zhitao Gong, Tris Warkentin, Ludovic Peran, Minh Giang, Clément Farabet, Oriol Vinyals, Jeff Dean, Koray Kavukcuoglu, Demis Hassabis, Zoubin Ghahramani, Douglas Eck, Joelle Barral, Fernando Pereira, Eli Collins, Armand Joulin, Noah Fiedel, Evan Senter, Alek Andreev, Kathleen Kenealy
|
2024-04-16T12:52:47Z
|
cs.CL, cs.AI
|
This work introduces Gemma, a family of lightweight, state-of-the art open
models built from the research and technology used to create Gemini models.
Gemma models demonstrate strong performance across academic benchmarks for
language understanding, reasoning, and safety. We release two sizes of models
(2 billion and 7 billion parameters), and provide both pretrained and
fine-tuned checkpoints. Gemma outperforms similarly sized open models on 11 out
of 18 text-based tasks, and we present comprehensive evaluations of safety and
responsibility aspects of the models, alongside a detailed description of model
development. We believe the responsible release of LLMs is critical for
improving the safety of frontier models, and for enabling the next wave of LLM
innovations.
|
cs.CL
|
Gemma: Open Models Based on Gemini Research and Technology
|
2024-03-13T06:54:47Z
|
http://arxiv.org/pdf/2403.08293v3
|
http://arxiv.org/abs/2403.08293v3
|
Xiang Hu, Pengyu Ji, Qingyang Zhu, Wei Wu, Kewei Tu
|
2024-06-17T16:22:52Z
|
cs.CL, cs.AI
|
A syntactic language model (SLM) incrementally generates a sentence with its
syntactic tree in a left-to-right manner. We present Generative Pretrained
Structured Transformers (GPST), an unsupervised SLM at scale capable of being
pre-trained from scratch on raw texts with high parallelism. GPST circumvents
the limitations of previous SLMs such as relying on gold trees and sequential
training. It consists of two components, a usual SLM supervised by a
uni-directional language modeling loss, and an additional composition model,
which induces syntactic parse trees and computes constituent representations,
supervised by a bi-directional language modeling loss. We propose a
representation surrogate to enable joint parallel training of the two models in
a hard-EM fashion. We pre-train GPST on OpenWebText, a corpus with $9$ billion
tokens, and demonstrate the superiority of GPST over GPT-2 with a comparable
size in numerous tasks covering both language understanding and language
generation. Meanwhile, GPST also significantly outperforms existing
unsupervised SLMs on left-to-right grammar induction, while holding a
substantial acceleration on training.
|
cs.CL
|
Generative Pretrained Structured Transformers: Unsupervised Syntactic Language Models at Scale
|
2024-03-13T06:18:48Z
|
http://arxiv.org/pdf/2403.08281v4
|
http://arxiv.org/abs/2403.08281v4
|
Ning Ding, Yulin Chen, Ganqu Cui, Xingtai Lv, Weilin Zhao, Ruobing Xie, Bowen Zhou, Zhiyuan Liu, Maosong Sun
|
2024-03-26T09:29:51Z
|
cs.CL, cs.AI
|
Underlying data distributions of natural language, programming code, and
mathematical symbols vary vastly, presenting a complex challenge for large
language models (LLMs) that strive to achieve high performance across all three
domains simultaneously. Achieving a very high level of proficiency for an LLM
within a specific domain often requires extensive training with relevant
corpora, which is typically accompanied by a sacrifice in performance in other
domains. In this paper, we propose to fuse models that are already
highly-specialized directly. The proposed fusing framework, UltraFuser,
consists of three distinct specialists that are already sufficiently trained on
language, coding, and mathematics. A token-level gating mechanism is introduced
to blend the specialists' outputs. A two-stage training strategy accompanied by
balanced sampling is designed to ensure stability. To effectively train the
fused model, we further construct a high-quality supervised instruction tuning
dataset, UltraChat 2, which includes text, code, and mathematical content. This
dataset comprises approximately 300,000 instructions and covers a wide range of
topics in each domain. Experiments show that our model could simultaneously
achieve mastery of the three crucial domains.
|
cs.CL
|
Mastering Text, Code and Math Simultaneously via Fusing Highly Specialized Language Models
|
2024-03-13T05:51:57Z
|
http://arxiv.org/pdf/2403.08272v1
|
http://arxiv.org/abs/2403.08272v1
|
Jieun Han, Haneul Yoo, Junho Myung, Minsun Kim, Tak Yeon Lee, So-Yeon Ahn, Alice Oh
|
2024-03-13T05:51:57Z
|
cs.CL
|
The integration of generative AI in education is expanding, yet empirical
analyses of large-scale and real-world interactions between students and AI
systems still remain limited. Addressing this gap, we present RECIPE4U (RECIPE
for University), a dataset sourced from a semester-long experiment with 212
college students in English as Foreign Language (EFL) writing courses. During
the study, students engaged in dialogues with ChatGPT to revise their essays.
RECIPE4U includes comprehensive records of these interactions, including
conversation logs, students' intent, students' self-rated satisfaction, and
students' essay edit histories. In particular, we annotate the students'
utterances in RECIPE4U with 13 intention labels based on our coding schemes. We
establish baseline results for two subtasks in task-oriented dialogue systems
within educational contexts: intent detection and satisfaction estimation. As a
foundational step, we explore student-ChatGPT interaction patterns through
RECIPE4U and analyze them by focusing on students' dialogue, essay data
statistics, and students' essay edits. We further illustrate potential
applications of RECIPE4U dataset for enhancing the incorporation of LLMs in
educational frameworks. RECIPE4U is publicly available at
https://zeunie.github.io/RECIPE4U/.
|
cs.CL
|
RECIPE4U: Student-ChatGPT Interaction Dataset in EFL Writing Education
|
2024-03-13T05:44:50Z
|
http://arxiv.org/pdf/2403.14683v1
|
http://arxiv.org/abs/2403.14683v1
|
Gokul Puthumanaillam, Manav Vora, Pranay Thangeda, Melkior Ornik
|
2024-03-13T05:44:50Z
|
cs.CY, cs.AI, cs.CL, cs.LG
|
This paper examines the challenges associated with achieving life-long
superalignment in AI systems, particularly large language models (LLMs).
Superalignment is a theoretical framework that aspires to ensure that
superintelligent AI systems act in accordance with human values and goals.
Despite its promising vision, we argue that achieving superalignment requires
substantial changes in the current LLM architectures due to their inherent
limitations in comprehending and adapting to the dynamic nature of these human
ethics and evolving global scenarios. We dissect the challenges of encoding an
ever-changing spectrum of human values into LLMs, highlighting the
discrepancies between static AI models and the dynamic nature of human
societies. To illustrate these challenges, we analyze two distinct examples:
one demonstrates a qualitative shift in human values, while the other presents
a quantifiable change. Through these examples, we illustrate how LLMs,
constrained by their training data, fail to align with contemporary human
values and scenarios. The paper concludes by exploring potential strategies to
address and possibly mitigate these alignment discrepancies, suggesting a path
forward in the pursuit of more adaptable and responsive AI systems.
|
cs.CY
|
A Moral Imperative: The Need for Continual Superalignment of Large Language Models
|
2024-03-13T05:38:39Z
|
http://arxiv.org/pdf/2403.15426v1
|
http://arxiv.org/abs/2403.15426v1
|
Zhangquan Chen, Chunjiang Liu, Haobin Duan
|
2024-03-13T05:38:39Z
|
cs.LG, cs.AI, cs.CL, I.2.7
|
In this paper, we propose an end-to-end prior-based three-phases supervised
fine-tuned model, which is proved more competitive than traditional fine-tuning
method. More specifically, our model realizes the structural disassembly and
incremental guided output of educational knowledge. To this end, we robustify
data classification of three types via a sampler and overlap estimation neural
network, and inject the preprocessing datasets into pre-trained model in three
batches for LORA fine-tuning. Then, we design a prior module couples system
prompt, vector databases, and abstract syntax tree task segmentation. Finally,
the compression method and regularization constraint are applied to the
prior-based fine-tuned model, followed by text filter at the output end to
obtain incremental guided results. Our model represents the first research
effort to truly embody the tutor role with the features of abundant educational
knowledge, step-by-step incremental guided outputs and non-disclosure of
answers. Extensive experiments report that our model also achieves
state-of-the-art in code abilities compared to open-source models, reaching an
impressive 75.10% on the HumanEval (@pass 1) benchmark. Additionally, our model
maintains strong conversational capabilities, with the 13B quantized version
achieving scores of 56.34, 50.60, and 45.27 respectively on the MMLU, C-Eval,
and AGIEval (5 shot) dialogue evaluation benchmarks.
|
cs.LG
|
A Three-Phases SFT Hybrid Model Integrated Strong Prior Module and Data Overlap Estimation in the Eduation Context
|
2024-03-13T05:20:45Z
|
http://arxiv.org/pdf/2403.08258v2
|
http://arxiv.org/abs/2403.08258v2
|
Wenjing Zhu, Sining Sun, Changhao Shan, Peng Fan, Qing Yang
|
2024-05-21T02:06:26Z
|
cs.CL, cs.LG
|
Conformer-based attention models have become the de facto backbone model for
Automatic Speech Recognition tasks. A blank symbol is usually introduced to
align the input and output sequences for CTC or RNN-T models. Unfortunately,
the long input length overloads computational budget and memory consumption
quadratically by attention mechanism. In this work, we propose a
"Skip-and-Recover" Conformer architecture, named Skipformer, to squeeze
sequence input length dynamically and inhomogeneously. Skipformer uses an
intermediate CTC output as criteria to split frames into three groups: crucial,
skipping and ignoring. The crucial group feeds into next conformer blocks and
its output joint with skipping group by original temporal order as the final
encoder output. Experiments show that our model reduces the input sequence
length by 31 times on Aishell-1 and 22 times on Librispeech corpus. Meanwhile,
the model can achieve better recognition accuracy and faster inference speed
than recent baseline models. Our code is open-sourced and available online.
|
cs.CL
|
Skipformer: A Skip-and-Recover Strategy for Efficient Speech Recognition
|
2024-03-13T04:14:33Z
|
http://arxiv.org/pdf/2403.08229v1
|
http://arxiv.org/abs/2403.08229v1
|
Zhenrong Cheng, Jiayan Guo, Hao Sun, Yan Zhang
|
2024-03-13T04:14:33Z
|
cs.CL
|
Current disfluency detection methods heavily rely on costly and scarce
human-annotated data. To tackle this issue, some approaches employ heuristic or
statistical features to generate disfluent sentences, partially improving
detection performance. However, these sentences often deviate from real-life
scenarios, constraining overall model enhancement. In this study, we propose a
lightweight data augmentation approach for disfluency detection, utilizing the
superior generative and semantic understanding capabilities of large language
model (LLM) to generate disfluent sentences as augmentation data. We leverage
LLM to generate diverse and more realistic sentences guided by specific
prompts, without the need for fine-tuning the LLM. Subsequently, we apply an
uncertainty-aware data filtering approach to improve the quality of the
generated sentences, utilized in training a small detection model for improved
performance. Experiments using enhanced data yielded state-of-the-art results.
The results showed that using a small amount of LLM-generated enhanced data can
significantly improve performance, thereby further enhancing
cost-effectiveness.
|
cs.CL
|
Boosting Disfluency Detection with Large Language Model as Disfluency Generator
|
2024-03-13T03:31:26Z
|
http://arxiv.org/pdf/2403.08217v1
|
http://arxiv.org/abs/2403.08217v1
|
Yichao Wu, Zhengyu Jin, Chenxi Shi, Penghao Liang, Tong Zhan
|
2024-03-13T03:31:26Z
|
cs.CL, cs.LG
|
This paper explores the application of deep learning techniques, particularly
focusing on BERT models, in sentiment analysis. It begins by introducing the
fundamental concept of sentiment analysis and how deep learning methods are
utilized in this domain. Subsequently, it delves into the architecture and
characteristics of BERT models. Through detailed explanation, it elucidates the
application effects and optimization strategies of BERT models in sentiment
analysis, supported by experimental validation. The experimental findings
indicate that BERT models exhibit robust performance in sentiment analysis
tasks, with notable enhancements post fine-tuning. Lastly, the paper concludes
by summarizing the potential applications of BERT models in sentiment analysis
and suggests directions for future research and practical implementations.
|
cs.CL
|
Research on the Application of Deep Learning-based BERT Model in Sentiment Analysis
|
2024-03-13T03:22:02Z
|
http://arxiv.org/pdf/2403.08213v1
|
http://arxiv.org/abs/2403.08213v1
|
Baixiang Huang, Canyu Chen, Kai Shu
|
2024-03-13T03:22:02Z
|
cs.CL
|
The ability to accurately identify authorship is crucial for verifying
content authenticity and mitigating misinformation. Large Language Models
(LLMs) have demonstrated exceptional capacity for reasoning and
problem-solving. However, their potential in authorship analysis, encompassing
authorship verification and attribution, remains underexplored. This paper
conducts a comprehensive evaluation of LLMs in these critical tasks.
Traditional studies have depended on hand-crafted stylistic features, whereas
state-of-the-art approaches leverage text embeddings from pre-trained language
models. These methods, which typically require fine-tuning on labeled data,
often suffer from performance degradation in cross-domain applications and
provide limited explainability. This work seeks to address three research
questions: (1) Can LLMs perform zero-shot, end-to-end authorship verification
effectively? (2) Are LLMs capable of accurately attributing authorship among
multiple candidates authors (e.g., 10 and 20)? (3) How can LLMs provide
explainability in authorship analysis, particularly through the role of
linguistic features? Moreover, we investigate the integration of explicit
linguistic features to guide LLMs in their reasoning processes. Our extensive
assessment demonstrates LLMs' proficiency in both tasks without the need for
domain-specific fine-tuning, providing insights into their decision-making via
a detailed analysis of linguistic features. This establishes a new benchmark
for future research on LLM-based authorship analysis. The code and data are
available at https://github.com/baixianghuang/authorship-llm.
|
cs.CL
|
Can Large Language Models Identify Authorship?
|
2024-03-13T03:15:05Z
|
http://arxiv.org/pdf/2403.08211v2
|
http://arxiv.org/abs/2403.08211v2
|
Liang Yao
|
2024-05-22T21:06:37Z
|
cs.CL, cs.AI
|
Prompting methods play a crucial role in enhancing the capabilities of
pre-trained large language models (LLMs). We explore how contrastive prompting
(CP) significantly improves the ability of large language models to perform
complex reasoning. We demonstrate that LLMs are decent contrastive reasoners by
simply adding "Let's give a correct and a wrong answer." before LLMs provide
answers. Experiments on various large language models show that zero-shot
contrastive prompting improves performance on a range of arithmetic,
commonsense, and symbolic reasoning tasks without any hand-crafted few-shot
examples, such as increasing the accuracy on GSM8K from 35.9% to 88.8% and
AQUA-RAT from 41.3% to 62.2% with the state-of-the-art GPT-4 model. Our method
not only surpasses zero-shot CoT and few-shot CoT in most arithmetic and
commonsense reasoning tasks but also can seamlessly integrate with existing
prompting methods, resulting in improved or comparable results when compared to
state-of-the-art methods. Our code is available at
https://github.com/yao8839836/cp
|
cs.CL
|
Large Language Models are Contrastive Reasoners
|
2024-03-13T02:53:36Z
|
http://arxiv.org/pdf/2403.13002v3
|
http://arxiv.org/abs/2403.13002v3
|
Shuo Jiang, Jianxi Luo
|
2024-05-23T02:19:12Z
|
cs.HC, cs.AI, cs.CL, I.2.7; I.2.1
|
Researchers and innovators have made enormous efforts in developing ideation
methods, such as morphological analysis and design-by-analogy, to aid
engineering design ideation for problem solving and innovation. Among these,
the Theory of Inventive Problem Solving (TRIZ) stands out as one of the most
well-known approaches, widely applied for systematic innovation. However, the
complexity of TRIZ resources and concepts, coupled with its reliance on users'
knowledge, experience, and reasoning capabilities, limits its practicality.
Therefore, we explore the recent advances of large language models (LLMs) for a
generative approach to bridge this gap. This paper proposes AutoTRIZ, an
artificial ideation tool that uses LLMs to automate and enhance the TRIZ
methodology. By leveraging the broad knowledge and advanced reasoning
capabilities of LLMs, AutoTRIZ offers a novel approach for design automation
and interpretable ideation with artificial intelligence. AutoTRIZ takes a
problem statement from the user as its initial input, and automatically
generates a solution report after the reasoning process. We demonstrate and
evaluate the effectiveness of AutoTRIZ through consistency experiments in
contradiction detection, and a case study comparing solutions generated by
AutoTRIZ with the experts' analyses from the textbook. Moreover, the proposed
LLM-based framework holds the potential for extension to automate other
knowledge-based ideation methods, including SCAMPER, Design Heuristics, and
Design-by-Analogy, paving the way for a new era of artificial ideation for
design innovation.
|
cs.HC
|
AutoTRIZ: Artificial Ideation with TRIZ and Large Language Models
|
2024-03-13T02:46:17Z
|
http://arxiv.org/pdf/2403.08198v1
|
http://arxiv.org/abs/2403.08198v1
|
Jonathan Dunn
|
2024-03-13T02:46:17Z
|
cs.CL
|
This paper investigates the impact of corpus creation decisions on large
multi-lingual geographic web corpora. Beginning with a 427 billion word corpus
derived from the Common Crawl, three methods are used to improve the quality of
sub-corpora representing specific language-country pairs like New Zealand
English: (i) the agreement of independent language identification systems, (ii)
hash-based deduplication, and (iii) location-specific outlier detection. The
impact of each of these steps is then evaluated at the language level and the
country level by using corpus similarity measures to compare each resulting
corpus with baseline data sets. The goal is to understand the impact of
upstream data cleaning decisions on downstream corpora with a specific focus on
under-represented languages and populations. The evaluation shows that the
validity of sub-corpora is improved with each stage of cleaning but that this
improvement is unevenly distributed across languages and populations. This
result shows how standard corpus creation techniques can accidentally exclude
under-represented populations.
|
cs.CL
|
Validating and Exploring Large Geographic Corpora
|
2024-03-13T02:41:53Z
|
http://arxiv.org/pdf/2403.08196v1
|
http://arxiv.org/abs/2403.08196v1
|
Jiayu Du, Jinpeng Li, Guoguo Chen, Wei-Qiang Zhang
|
2024-03-13T02:41:53Z
|
cs.CL, eess.AS
|
In the wake of the surging tide of deep learning over the past decade,
Automatic Speech Recognition (ASR) has garnered substantial attention, leading
to the emergence of numerous publicly accessible ASR systems that are actively
being integrated into our daily lives. Nonetheless, the impartial and
replicable evaluation of these ASR systems encounters challenges due to various
crucial subtleties. In this paper we introduce the SpeechColab Leaderboard, a
general-purpose, open-source platform designed for ASR evaluation. With this
platform: (i) We report a comprehensive benchmark, unveiling the current
state-of-the-art panorama for ASR systems, covering both open-source models and
industrial commercial services. (ii) We quantize how distinct nuances in the
scoring pipeline influence the final benchmark outcomes. These include nuances
related to capitalization, punctuation, interjection, contraction, synonym
usage, compound words, etc. These issues have gained prominence in the context
of the transition towards an End-to-End future. (iii) We propose a practical
modification to the conventional Token-Error-Rate (TER) evaluation metric, with
inspirations from Kolmogorov complexity and Normalized Information Distance
(NID). This adaptation, called modified-TER (mTER), achieves proper
normalization and symmetrical treatment of reference and hypothesis. By
leveraging this platform as a large-scale testing ground, this study
demonstrates the robustness and backward compatibility of mTER when compared to
TER. The SpeechColab Leaderboard is accessible at
https://github.com/SpeechColab/Leaderboard
|
cs.CL
|
SpeechColab Leaderboard: An Open-Source Platform for Automatic Speech Recognition Evaluation
|
2024-03-13T02:32:41Z
|
http://arxiv.org/pdf/2403.09732v4
|
http://arxiv.org/abs/2403.09732v4
|
Zhishuai Li, Xiang Wang, Jingjing Zhao, Sun Yang, Guoqing Du, Xiaoru Hu, Bin Zhang, Yuxiao Ye, Ziyue Li, Rui Zhao, Hangyu Mao
|
2024-06-02T02:58:53Z
|
cs.CL, cs.AI
|
Recent advancements in Text-to-SQL (Text2SQL) emphasize stimulating the large
language models (LLM) on in-context learning, achieving significant results.
Nevertheless, they face challenges when dealing with verbose database
information and complex user intentions. This paper presents a two-stage
framework to enhance the performance of current LLM-based natural language to
SQL systems. We first introduce a novel prompt representation, called
reference-enhanced representation, which includes schema information and
randomly sampled cell values from tables to instruct LLMs in generating SQL
queries. Then, in the first stage, question-SQL pairs are retrieved as few-shot
demonstrations, prompting the LLM to generate a preliminary SQL (PreSQL). After
that, the mentioned entities in PreSQL are parsed to conduct schema linking,
which can significantly compact the useful information. In the second stage,
with the linked schema, we simplify the prompt's schema information and
instruct the LLM to produce the final SQL. Finally, as the post-refinement
module, we propose using cross-consistency across different LLMs rather than
self-consistency within a particular LLM. Our methods achieve new SOTA results
on the Spider benchmark, with an execution accuracy of 87.6%.
|
cs.CL
|
PET-SQL: A Prompt-Enhanced Two-Round Refinement of Text-to-SQL with Cross-consistency
|
2024-03-13T02:26:16Z
|
http://arxiv.org/pdf/2403.08192v1
|
http://arxiv.org/abs/2403.08192v1
|
Xingyu Lu, He Cao, Zijing Liu, Shengyuan Bai, Leqing Chen, Yuan Yao, Hai-Tao Zheng, Yu Li
|
2024-03-13T02:26:16Z
|
cs.CL, q-bio.BM
|
Large language models are playing an increasingly significant role in
molecular research, yet existing models often generate erroneous information,
posing challenges to accurate molecular comprehension. Traditional evaluation
metrics for generated content fail to assess a model's accuracy in molecular
understanding. To rectify the absence of factual evaluation, we present
MoleculeQA, a novel question answering (QA) dataset which possesses 62K QA
pairs over 23K molecules. Each QA pair, composed of a manual question, a
positive option and three negative options, has consistent semantics with a
molecular description from authoritative molecular corpus. MoleculeQA is not
only the first benchmark for molecular factual bias evaluation but also the
largest QA dataset for molecular research. A comprehensive evaluation on
MoleculeQA for existing molecular LLMs exposes their deficiencies in specific
areas and pinpoints several particularly crucial factors for molecular
understanding.
|
cs.CL
|
MoleculeQA: A Dataset to Evaluate Factual Accuracy in Molecular Comprehension
|
2024-03-13T02:23:13Z
|
http://arxiv.org/pdf/2403.08189v1
|
http://arxiv.org/abs/2403.08189v1
|
Changbing Yang, Garrett Nicolai, Miikka Silfverberg
|
2024-03-13T02:23:13Z
|
cs.CL
|
We investigate automatic interlinear glossing in low-resource settings. We
augment a hard-attentional neural model with embedded translation information
extracted from interlinear glossed text. After encoding these translations
using large language models, specifically BERT and T5, we introduce a
character-level decoder for generating glossed output. Aided by these
enhancements, our model demonstrates an average improvement of 3.97\%-points
over the previous state of the art on datasets from the SIGMORPHON 2023 Shared
Task on Interlinear Glossing. In a simulated ultra low-resource setting,
trained on as few as 100 sentences, our system achieves an average 9.78\%-point
improvement over the plain hard-attentional baseline. These results highlight
the critical role of translation information in boosting the system's
performance, especially in processing and interpreting modest data sources. Our
findings suggest a promising avenue for the documentation and preservation of
languages, with our experiments on shared task datasets indicating significant
advancements over the existing state of the art.
|
cs.CL
|
Embedded Translations for Low-resource Automated Glossing
|
2024-03-13T02:20:05Z
|
http://arxiv.org/pdf/2403.08187v1
|
http://arxiv.org/abs/2403.08187v1
|
Taekyung Ahn, Yeonjung Hong, Younggon Im, Do Hyung Kim, Dayoung Kang, Joo Won Jeong, Jae Won Kim, Min Jung Kim, Ah-ra Cho, Dae-Hyun Jang, Hosung Nam
|
2024-03-13T02:20:05Z
|
cs.CL, cs.SD, eess.AS, I.2.7
|
This study presents a model of automatic speech recognition (ASR) designed to
diagnose pronunciation issues in children with speech sound disorders (SSDs) to
replace manual transcriptions in clinical procedures. Since ASR models trained
for general purposes primarily predict input speech into real words, employing
a well-known high-performance ASR model for evaluating pronunciation in
children with SSDs is impractical. We fine-tuned the wav2vec 2.0 XLS-R model to
recognize speech as pronounced rather than as existing words. The model was
fine-tuned with a speech dataset from 137 children with inadequate speech
production pronouncing 73 Korean words selected for actual clinical diagnosis.
The model's predictions of the pronunciations of the words matched the human
annotations with about 90% accuracy. While the model still requires improvement
in recognizing unclear pronunciation, this study demonstrates that ASR models
can streamline complex pronunciation error diagnostic procedures in clinical
fields.
|
cs.CL
|
Automatic Speech Recognition (ASR) for the Diagnosis of pronunciation of Speech Sound Disorders in Korean children
|
2024-03-13T01:56:32Z
|
http://arxiv.org/pdf/2403.08174v1
|
http://arxiv.org/abs/2403.08174v1
|
Yuta Mukobara, Yutaro Shigeto, Masashi Shimbo
|
2024-03-13T01:56:32Z
|
cs.CL, cs.AI
|
We explore loss functions for fact verification in the FEVER shared task.
While the cross-entropy loss is a standard objective for training verdict
predictors, it fails to capture the heterogeneity among the FEVER verdict
classes. In this paper, we develop two task-specific objectives tailored to
FEVER. Experimental results confirm that the proposed objective functions
outperform the standard cross-entropy. Performance is further improved when
these objectives are combined with simple class weighting, which effectively
overcomes the imbalance in the training data. The souce code is available at
https://github.com/yuta-mukobara/RLF-KGAT
|
cs.CL
|
Rethinking Loss Functions for Fact Verification
|
2024-03-13T01:38:42Z
|
http://arxiv.org/pdf/2403.08167v2
|
http://arxiv.org/abs/2403.08167v2
|
Teng Xiao, Chao Cui, Huaisheng Zhu, Vasant G. Honavar
|
2024-04-03T01:00:53Z
|
cs.LG, cs.CL, q-bio.QM
|
Recent advancements in biology and chemistry have leveraged multi-modal
learning, integrating molecules and their natural language descriptions to
enhance drug discovery. However, current pre-training frameworks are limited to
two modalities, and designing a unified network to process different modalities
(e.g., natural language, 2D molecular graphs, 3D molecular conformations, and
3D proteins) remains challenging due to inherent gaps among them. In this work,
we propose MolBind, a framework that trains encoders for multiple modalities
through contrastive learning, mapping all modalities to a shared feature space
for multi-modal semantic alignment. To facilitate effective pre-training of
MolBind on multiple modalities, we also build and collect a high-quality
dataset with four modalities, MolBind-M4, including graph-language,
conformation-language, graph-conformation, and conformation-protein paired
data. MolBind shows superior zero-shot learning performance across a wide range
of tasks, demonstrating its strong capability of capturing the underlying
semantics of multiple modalities.
|
cs.LG
|
MolBind: Multimodal Alignment of Language, Molecules, and Proteins
|
2024-03-12T23:59:15Z
|
http://arxiv.org/pdf/2403.08140v2
|
http://arxiv.org/abs/2403.08140v2
|
Shikhar Murty, Christopher Manning, Peter Shaw, Mandar Joshi, Kenton Lee
|
2024-06-09T01:12:26Z
|
cs.CL
|
Following natural language instructions by executing actions in digital
environments (e.g. web-browsers and REST APIs) is a challenging task for
language model (LM) agents. Unfortunately, LM agents often fail to generalize
to new environments without human demonstrations. This work presents BAGEL, a
method for bootstrapping LM agents without human supervision. BAGEL converts a
seed set of randomly explored trajectories or synthetic instructions, into
demonstrations, via round-trips between two noisy LM components: an LM labeler
which converts a trajectory into a synthetic instruction, and a zero-shot LM
agent which maps the synthetic instruction into a refined trajectory. By
performing these round-trips iteratively, BAGEL quickly converts the initial
distribution of trajectories towards those that are well-described by natural
language. We use BAGEL demonstrations to adapt a zero shot LM agent at test
time via in-context learning over retrieved demonstrations, and find
improvements of over 2-13% absolute on ToolQA and MiniWob++, with up to 13x
reduction in execution failures.
|
cs.CL
|
BAGEL: Bootstrapping Agents by Guiding Exploration with Language
|
2024-03-12T23:47:28Z
|
http://arxiv.org/pdf/2403.08137v1
|
http://arxiv.org/abs/2403.08137v1
|
Donghoon Shin, Lucy Lu Wang, Gary Hsieh
|
2024-03-12T23:47:28Z
|
cs.HC, cs.AI, cs.CL, H.5.2; I.2.7
|
Communicating design implications is common within the HCI community when
publishing academic papers, yet these papers are rarely read and used by
designers. One solution is to use design cards as a form of translational
resource that communicates valuable insights from papers in a more digestible
and accessible format to assist in design processes. However, creating design
cards can be time-consuming, and authors may lack the resources/know-how to
produce cards. Through an iterative design process, we built a system that
helps create design cards from academic papers using an LLM and text-to-image
model. Our evaluation with designers (N=21) and authors of selected papers
(N=12) revealed that designers perceived the design implications from our
design cards as more inspiring and generative, compared to reading original
paper texts, and the authors viewed our system as an effective way of
communicating their design implications. We also propose future enhancements
for AI-generated design cards.
|
cs.HC
|
From Paper to Card: Transforming Design Implications with Generative AI
|
2024-03-12T22:53:32Z
|
http://arxiv.org/pdf/2403.08115v2
|
http://arxiv.org/abs/2403.08115v2
|
Vincent Freiberger, Erik Buchmann
|
2024-05-08T14:47:39Z
|
cs.CY, cs.AI, cs.CL, K.4.m
|
Privacy policies are expected to inform data subjects about their data
protection rights and should explain the data controller's data management
practices. Privacy policies only fulfill their purpose, if they are correctly
interpreted, understood, and trusted by the data subject. This implies that a
privacy policy is written in a fair way, e.g., it does not use polarizing
terms, does not require a certain education, or does not assume a particular
social background. We outline our approach to assessing fairness in privacy
policies. We identify from fundamental legal sources and fairness research, how
the dimensions informational fairness, representational fairness and ethics /
morality are related to privacy policies. We propose options to automatically
assess policies in these fairness dimensions, based on text statistics,
linguistic methods and artificial intelligence. We conduct initial experiments
with German privacy policies to provide evidence that our approach is
applicable. Our experiments indicate that there are issues in all three
dimensions of fairness. This is important, as future privacy policies may be
used in a corpus for legal artificial intelligence models.
|
cs.CY
|
Legally Binding but Unfair? Towards Assessing Fairness of Privacy Policies
|
2024-03-12T22:36:27Z
|
http://arxiv.org/pdf/2403.08111v1
|
http://arxiv.org/abs/2403.08111v1
|
Ruican Zhong, Donghoon Shin, Rosemary Meza, Predrag Klasnja, Lucas Colusso, Gary Hsieh
|
2024-03-12T22:36:27Z
|
cs.HC, cs.AI, cs.CL, H.5.2; I.2.7
|
This paper explores the integration of causal pathway diagrams (CPD) into
human-centered design (HCD), investigating how these diagrams can enhance the
early stages of the design process. A dedicated CPD plugin for the online
collaborative whiteboard platform Miro was developed to streamline diagram
creation and offer real-time AI-driven guidance. Through a user study with
designers (N=20), we found that CPD's branching and its emphasis on causal
connections supported both divergent and convergent processes during design.
CPD can also facilitate communication among stakeholders. Additionally, we
found our plugin significantly reduces designers' cognitive workload and
increases their creativity during brainstorming, highlighting the implications
of AI-assisted tools in supporting creative work and evidence-based designs.
|
cs.HC
|
AI-Assisted Causal Pathway Diagram for Human-Centered Design
|
2024-03-12T22:23:08Z
|
http://arxiv.org/pdf/2403.08103v2
|
http://arxiv.org/abs/2403.08103v2
|
Ruslan Musaev
|
2024-03-14T10:47:32Z
|
cs.CL, cs.AI
|
In the age of information abundance, the ability to provide users with
contextually relevant and concise information is crucial. Keyword in Context
(KIC) generation is a task that plays a vital role in and generation
applications, such as search engines, personal assistants, and content
summarization. In this paper, we present a novel approach to generating
unambiguous and brief sentence-contexts for given keywords using the T5
transformer model, leveraging data obtained from the Context-Reverso API. The
code is available at https://github.com/Rusamus/word2context/tree/main .
|
cs.CL
|
Contextual Clarity: Generating Sentences with Transformer Models using Context-Reverso Data
|
2024-03-12T21:54:34Z
|
http://arxiv.org/pdf/2403.09728v1
|
http://arxiv.org/abs/2403.09728v1
|
Michael Rizvi, Maude Lizaire, Clara Lacroce, Guillaume Rabusseau
|
2024-03-12T21:54:34Z
|
cs.CL, cs.AI, cs.CC
|
Transformers are ubiquitous models in the natural language processing (NLP)
community and have shown impressive empirical successes in the past few years.
However, little is understood about how they reason and the limits of their
computational capabilities. These models do not process data sequentially, and
yet outperform sequential neural models such as RNNs. Recent work has shown
that these models can compactly simulate the sequential reasoning abilities of
deterministic finite automata (DFAs). This leads to the following question: can
transformers simulate the reasoning of more complex finite state machines? In
this work, we show that transformers can simulate weighted finite automata
(WFAs), a class of models which subsumes DFAs, as well as weighted tree
automata (WTA), a generalization of weighted automata to tree structured
inputs. We prove these claims formally and provide upper bounds on the sizes of
the transformer models needed as a function of the number of states the target
automata. Empirically, we perform synthetic experiments showing that
transformers are able to learn these compact solutions via standard
gradient-based training.
|
cs.CL
|
Simulating Weighted Automata over Sequences and Trees with Transformers
|
2024-03-12T21:15:38Z
|
http://arxiv.org/pdf/2403.08081v1
|
http://arxiv.org/abs/2403.08081v1
|
Yingcong Li, Yixiao Huang, M. Emrullah Ildiz, Ankit Singh Rawat, Samet Oymak
|
2024-03-12T21:15:38Z
|
cs.LG, cs.AI, cs.CL, math.OC
|
Transformer-based language models are trained on large datasets to predict
the next token given an input sequence. Despite this simple training objective,
they have led to revolutionary advances in natural language processing.
Underlying this success is the self-attention mechanism. In this work, we ask:
$\textit{What}$ $\textit{does}$ $\textit{a}$ $\textit{single}$
$\textit{self-attention}$ $\textit{layer}$ $\textit{learn}$ $\textit{from}$
$\textit{next-token}$ $\textit{prediction?}$ We show that training
self-attention with gradient descent learns an automaton which generates the
next token in two distinct steps: $\textbf{(1)}$ $\textbf{Hard}$
$\textbf{retrieval:}$ Given input sequence, self-attention precisely selects
the $\textit{high-priority}$ $\textit{input}$ $\textit{tokens}$ associated with
the last input token. $\textbf{(2)}$ $\textbf{Soft}$ $\textbf{composition:}$ It
then creates a convex combination of the high-priority tokens from which the
next token can be sampled. Under suitable conditions, we rigorously
characterize these mechanics through a directed graph over tokens extracted
from the training data. We prove that gradient descent implicitly discovers the
strongly-connected components (SCC) of this graph and self-attention learns to
retrieve the tokens that belong to the highest-priority SCC available in the
context window. Our theory relies on decomposing the model weights into a
directional component and a finite component that correspond to hard retrieval
and soft composition steps respectively. This also formalizes a related
implicit bias formula conjectured in [Tarzanagh et al. 2023]. We hope that
these findings shed light on how self-attention processes sequential data and
pave the path toward demystifying more complex architectures.
|
cs.LG
|
Mechanics of Next Token Prediction with Self-Attention
|
2024-03-12T21:06:31Z
|
http://arxiv.org/pdf/2403.09727v1
|
http://arxiv.org/abs/2403.09727v1
|
Robert Lakatos, Peter Pollner, Andras Hajdu, Tamas Joo
|
2024-03-12T21:06:31Z
|
cs.CL, cs.AI, cs.LG
|
The development of generative large language models (G-LLM) opened up new
opportunities for the development of new types of knowledge-based systems
similar to ChatGPT, Bing, or Gemini. Fine-tuning (FN) and Retrieval-Augmented
Generation (RAG) are the techniques that can be used to implement domain
adaptation for the development of G-LLM-based knowledge systems. In our study,
using ROUGE, BLEU, METEOR scores, and cosine similarity, we compare and examine
the performance of RAG and FN for the GPT-J-6B, OPT-6.7B, LlaMA, LlaMA-2
language models. Based on measurements shown on different datasets, we
demonstrate that RAG-based constructions are more efficient than models
produced with FN. We point out that connecting RAG and FN is not trivial,
because connecting FN models with RAG can cause a decrease in performance.
Furthermore, we outline a simple RAG-based architecture which, on average,
outperforms the FN models by 16% in terms of the ROGUE score, 15% in the case
of the BLEU score, and 53% based on the cosine similarity. This shows the
significant advantage of RAG over FN in terms of hallucination, which is not
offset by the fact that the average 8% better METEOR score of FN models
indicates greater creativity compared to RAG.
|
cs.CL
|
Investigating the performance of Retrieval-Augmented Generation and fine-tuning for the development of AI-driven knowledge-based systems
|
2024-03-12T20:11:38Z
|
http://arxiv.org/pdf/2403.08059v2
|
http://arxiv.org/abs/2403.08059v2
|
Benjamin D. Killeen, Liam J. Wang, Han Zhang, Mehran Armand, Russell H. Taylor, Dave Dreizin, Greg Osgood, Mathias Unberath
|
2024-03-28T00:59:37Z
|
cs.CV, cs.AI, cs.CL, cs.LG
|
Automated X-ray image segmentation would accelerate research and development
in diagnostic and interventional precision medicine. Prior efforts have
contributed task-specific models capable of solving specific image analysis
problems, but the utility of these models is restricted to their particular
task domain, and expanding to broader use requires additional data, labels, and
retraining efforts. Recently, foundation models (FMs) -- machine learning
models trained on large amounts of highly variable data thus enabling broad
applicability -- have emerged as promising tools for automated image analysis.
Existing FMs for medical image analysis focus on scenarios and modalities where
objects are clearly defined by visually apparent boundaries, such as surgical
tool segmentation in endoscopy. X-ray imaging, by contrast, does not generally
offer such clearly delineated boundaries or structure priors. During X-ray
image formation, complex 3D structures are projected in transmission onto the
imaging plane, resulting in overlapping features of varying opacity and shape.
To pave the way toward an FM for comprehensive and automated analysis of
arbitrary medical X-ray images, we develop FluoroSAM, a language-aligned
variant of the Segment-Anything Model, trained from scratch on 1.6M synthetic
X-ray images. FluoroSAM is trained on data including masks for 128 organ types
and 464 non-anatomical objects, such as tools and implants. In real X-ray
images of cadaveric specimens, FluoroSAM is able to segment bony anatomical
structures based on text-only prompting with 0.51 and 0.79 DICE with
point-based refinement, outperforming competing SAM variants for all
structures. FluoroSAM is also capable of zero-shot generalization to segmenting
classes beyond the training set thanks to its language alignment, which we
demonstrate for full lung segmentation on real chest X-rays.
|
cs.CV
|
FluoroSAM: A Language-aligned Foundation Model for X-ray Image Segmentation
|
2024-03-12T20:10:04Z
|
http://arxiv.org/pdf/2403.08058v2
|
http://arxiv.org/abs/2403.08058v2
|
Saurabh Agarwal, Bilge Acun, Basil Hosmer, Mostafa Elhoushi, Yejin Lee, Shivaram Venkataraman, Dimitris Papailiopoulos, Carole-Jean Wu
|
2024-04-27T22:45:39Z
|
cs.LG, cs.CL
|
Large Language Models (LLMs) with hundreds of billions of parameters have
transformed the field of machine learning. However, serving these models at
inference time is both compute and memory intensive, where a single request can
require multiple GPUs and tens of Gigabytes of memory. Multi-Head Attention is
one of the key components of LLMs, which can account for over 50% of LLMs
memory and compute requirement. We observe that there is a high amount of
redundancy across heads on which tokens they pay attention to. Based on this
insight, we propose Clustered Head Attention (CHAI). CHAI combines heads with a
high amount of correlation for self-attention at runtime, thus reducing both
memory and compute. In our experiments, we show that CHAI is able to reduce the
memory requirements for storing K,V cache by up to 21.4% and inference time
latency by up to 1.73x without any fine-tuning required. CHAI achieves this
with a maximum 3.2% deviation in accuracy across 3 different models (i.e.
OPT-66B, LLAMA-7B, LLAMA-33B) and 5 different evaluation datasets.
|
cs.LG
|
CHAI: Clustered Head Attention for Efficient LLM Inference
|
2024-03-12T19:57:39Z
|
http://arxiv.org/pdf/2403.08053v1
|
http://arxiv.org/abs/2403.08053v1
|
Anmol Singhal, Chirag Jain, Preethu Rose Anish, Arkajyoti Chakraborty, Smita Ghaisas
|
2024-03-12T19:57:39Z
|
cs.CL
|
Enterprises frequently enter into commercial contracts that can serve as
vital sources of project-specific requirements. Contractual clauses are
obligatory, and the requirements derived from contracts can detail the
downstream implementation activities that non-legal stakeholders, including
requirement analysts, engineers, and delivery personnel, need to conduct.
However, comprehending contracts is cognitively demanding and error-prone for
such stakeholders due to the extensive use of Legalese and the inherent
complexity of contract language. Furthermore, contracts often contain
ambiguously worded clauses to ensure comprehensive coverage. In contrast,
non-legal stakeholders require a detailed and unambiguous comprehension of
contractual clauses to craft actionable requirements. In this work, we
introduce a novel legal NLP task that involves generating clarification
questions for contracts. These questions aim to identify contract ambiguities
on a document level, thereby assisting non-legal stakeholders in obtaining the
necessary details for eliciting requirements. This task is challenged by three
core issues: (1) data availability, (2) the length and unstructured nature of
contracts, and (3) the complexity of legal text. To address these issues, we
propose ConRAP, a retrieval-augmented prompting framework for generating
clarification questions to disambiguate contractual text. Experiments conducted
on contracts sourced from the publicly available CUAD dataset show that ConRAP
with ChatGPT can detect ambiguities with an F2 score of 0.87. 70% of the
generated clarification questions are deemed useful by human evaluators.
|
cs.CL
|
Generating Clarification Questions for Disambiguating Contracts
|
2024-03-12T19:40:18Z
|
http://arxiv.org/pdf/2403.08046v1
|
http://arxiv.org/abs/2403.08046v1
|
Charlie Campanella, Rob van der Goot
|
2024-03-12T19:40:18Z
|
cs.CL, I.2.7
|
Large language models (LLMs) have emerged as a useful technology for job
matching, for both candidates and employers. Job matching is often based on a
particular geographic location, such as a city or region. However, LLMs have
known biases, commonly derived from their training data. In this work, we aim
to quantify the metropolitan size bias encoded within large language models,
evaluating zero-shot salary, employer presence, and commute duration
predictions in 384 of the United States' metropolitan regions. Across all
benchmarks, we observe negative correlations between the metropolitan size and
the performance of the LLMS, indicating that smaller regions are indeed
underrepresented. More concretely, the smallest 10 metropolitan regions show
upwards of 300% worse benchmark performance than the largest 10.
|
cs.CL
|
Big City Bias: Evaluating the Impact of Metropolitan Size on Computational Job Market Abilities of Language Models
|
2024-03-12T19:34:54Z
|
http://arxiv.org/pdf/2403.08043v1
|
http://arxiv.org/abs/2403.08043v1
|
Shuai Liu, Shantanu Agarwal, Jonathan May
|
2024-03-12T19:34:54Z
|
cs.CL
|
Authorship style transfer aims to rewrite a given text into a specified
target while preserving the original meaning in the source. Existing approaches
rely on the availability of a large number of target style exemplars for model
training. However, these overlook cases where a limited number of target style
examples are available. The development of parameter-efficient transfer
learning techniques and policy optimization (PO) approaches suggest lightweight
PO is a feasible approach to low-resource style transfer. In this work, we
propose a simple two step tune-and-optimize technique for low-resource textual
style transfer. We apply our technique to authorship transfer as well as a
larger-data native language style task and in both cases find it outperforms
state-of-the-art baseline models.
|
cs.CL
|
Authorship Style Transfer with Policy Optimization
|
2024-03-12T19:12:28Z
|
http://arxiv.org/pdf/2403.08035v1
|
http://arxiv.org/abs/2403.08035v1
|
Tharindu Kumarage, Amrita Bhattacharjee, Joshua Garland
|
2024-03-12T19:12:28Z
|
cs.CL, cs.AI
|
Large language models (LLMs) excel in many diverse applications beyond
language generation, e.g., translation, summarization, and sentiment analysis.
One intriguing application is in text classification. This becomes pertinent in
the realm of identifying hateful or toxic speech -- a domain fraught with
challenges and ethical dilemmas. In our study, we have two objectives: firstly,
to offer a literature review revolving around LLMs as classifiers, emphasizing
their role in detecting and classifying hateful or toxic content. Subsequently,
we explore the efficacy of several LLMs in classifying hate speech: identifying
which LLMs excel in this task as well as their underlying attributes and
training. Providing insight into the factors that contribute to an LLM
proficiency (or lack thereof) in discerning hateful content. By combining a
comprehensive literature review with an empirical analysis, our paper strives
to shed light on the capabilities and constraints of LLMs in the crucial domain
of hate speech detection.
|
cs.CL
|
Harnessing Artificial Intelligence to Combat Online Hate: Exploring the Challenges and Opportunities of Large Language Models in Hate Speech Detection
|
2024-03-12T18:21:20Z
|
http://arxiv.org/pdf/2403.08011v1
|
http://arxiv.org/abs/2403.08011v1
|
Yash Sharma, Basil Abraham, Preethi Jyothi
|
2024-03-12T18:21:20Z
|
cs.CL, cs.AI, cs.LG
|
An important and difficult task in code-switched speech recognition is to
recognize the language, as lots of words in two languages can sound similar,
especially in some accents. We focus on improving performance of end-to-end
Automatic Speech Recognition models by conditioning transformer layers on
language ID of words and character in the output in an per layer supervised
manner. To this end, we propose two methods of introducing language specific
parameters and explainability in the multi-head attention mechanism, and
implement a Temporal Loss that helps maintain continuity in input alignment.
Despite being unable to reduce WER significantly, our method shows promise in
predicting the correct language from just spoken data. We introduce
regularization in the language prediction by dropping LID in the sequence,
which helps align long repeated output sequences.
|
cs.CL
|
Gujarati-English Code-Switching Speech Recognition using ensemble prediction of spoken language
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.