PyTorchKR Papers

최신 논문

총 1,009편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2504.05506공식 코드 · Papers with Code

ChartQAPro: A More Diverse and Challenging Benchmark for Chart Question Answering

Charts are ubiquitous, as people often use them to analyze data, answer questions, and discover critical insights. However, performing complex analytical tasks with charts requires significant perceptual and cognitive effort. Chart Question Answering (CQA) systems automate this process by enabling models to interpret and reason with visual representations of data. However, existing benchmarks like ChartQA lack real-world diversity and have recently shown performance saturation with modern large vision-language models (LVLMs). To address these limitations, we introduce ChartQAPro, a new benchmark that includes 1,341 charts from 157 diverse sources, spanning various chart types, including infographics and dashboards, and featuring 1,948 questions in various types, such as multiple-choice, conversational, hypothetical, and unanswerable questions, to better reflect real-world challenges. Our evaluations with 21 models show a substantial performance drop for LVLMs on ChartQAPro; e.g., Claude Sonnet 3.5 scores 90.5% on ChartQA but only 55.81% on ChartQAPro, underscoring the complexity of chart reasoning. We complement our findings with detailed error analyses and ablation studies, identifying key challenges and opportunities for advancing LVLMs in chart understanding and reasoning. We release ChartQAPro at https://github.com/vis-nlp/ChartQAPro.

arXiv:2504.07072

Kaleidoscope: In-language Exams for Massively Multilingual Vision Evaluation

The evaluation of vision-language models (VLMs) has mainly relied on English-language benchmarks, leaving significant gaps in both multilingual and multicultural coverage. While multilingual benchmarks have expanded, both in size and languages, many rely on translations of English datasets, failing to capture cultural nuances. In this work, we propose Kaleidoscope, as the most comprehensive exam benchmark to date for the multilingual evaluation of vision-language models. Kaleidoscope is a large-scale, in-language multimodal benchmark designed to evaluate VLMs across diverse languages and visual inputs. Kaleidoscope covers 18 languages and 14 different subjects, amounting to a total of 20,911 multiple-choice questions. Built through an open science collaboration with a diverse group of researchers worldwide, Kaleidoscope ensures linguistic and cultural authenticity. We evaluate top-performing multilingual vision-language models and find that they perform poorly on low-resource languages and in complex multimodal scenarios. Our results highlight the need for progress on culturally inclusive multimodal evaluation frameworks.

arXiv:2504.05040

InstructionBench: An Instructional Video Understanding Benchmark

Despite progress in video large language models (Video-LLMs), research on instructional video understanding, crucial for enhancing access to instructional content, remains insufficient. To address this, we introduce InstructionBench, an Instructional video understanding Benchmark, which challenges models' advanced temporal reasoning within instructional videos characterized by their strict step-by-step flow. Employing GPT-4, we formulate Q\&A pairs in open-ended and multiple-choice formats to assess both Coarse-Grained event-level and Fine-Grained object-level reasoning. Our filtering strategies exclude questions answerable purely by common-sense knowledge, focusing on visual perception and analysis when evaluating Video-LLM models. The benchmark finally contains 5k questions across over 700 videos. We evaluate the latest Video-LLMs on our InstructionBench, finding that closed-source models outperform open-source ones. However, even the best model, GPT-4o, achieves only 53.42\% accuracy, indicating significant gaps in temporal reasoning. To advance the field, we also develop a comprehensive instructional video dataset with over 19k Q\&A pairs from nearly 2.5k videos, using an automated data generation framework, thereby enriching the community's research resources.

arXiv:2504.03219

From ChatGPT to DeepSeek AI: A Comprehensive Analysis of Evolution, Deviation, and Future Implications in AI-Language Models

The rapid advancement of artificial intelligence (AI) has reshaped the field of natural language processing (NLP), with models like OpenAI ChatGPT and DeepSeek AI. Although ChatGPT established a strong foundation for conversational AI, DeepSeek AI introduces significant improvements in architecture, performance, and ethical considerations. This paper presents a detailed analysis of the evolution from ChatGPT to DeepSeek AI, highlighting their technical differences, practical applications, and broader implications for AI development. To assess their capabilities, we conducted a case study using a predefined set of multiple choice questions in various domains, evaluating the strengths and limitations of each model. By examining these aspects, we provide valuable insight into the future trajectory of AI, its potential to transform industries, and key research directions for improving AI-driven language models.

arXiv:2504.02227공식 코드 · Papers with Code

VEGAS: Towards Visually Explainable and Grounded Artificial Social Intelligence

Social Intelligence Queries (Social-IQ) serve as the primary multimodal benchmark for evaluating a model's social intelligence level. While impressive multiple-choice question(MCQ) accuracy is achieved by current solutions, increasing evidence shows that they are largely, and in some cases entirely, dependent on language modality, overlooking visual context. Additionally, the closed-set nature further prevents the exploration of whether and to what extent the reasoning path behind selection is correct. To address these limitations, we propose the Visually Explainable and Grounded Artificial Social Intelligence (VEGAS) model. As a generative multimodal model, VEGAS leverages open-ended answering to provide explainable responses, which enhances the clarity and evaluation of reasoning paths. To enable visually grounded answering, we propose a novel sampling strategy to provide the model with more relevant visual frames. We then enhance the model's interpretation of these frames through Generalist Instruction Fine-Tuning (GIFT), which aims to: i) learn multimodal-language transformations for fundamental emotional social traits, and ii) establish multimodal joint reasoning capabilities. Extensive experiments, comprising modality ablation, open-ended assessments, and supervised MCQ evaluations, consistently show that VEGAS effectively utilizes visual information in reasoning to produce correct and also credible answers. We expect this work to of fer a new perspective on Social-IQ and advance the development of human-like social AI.

arXiv:2503.23523공식 코드 · Papers with Code

Question-Aware Knowledge Graph Prompting for Enhancing Large Language Models

Large Language Models (LLMs) often struggle with tasks requiring external knowledge, such as knowledge-intensive Multiple Choice Question Answering (MCQA). Integrating Knowledge Graphs (KGs) can enhance reasoning; however, existing methods typically demand costly fine-tuning or retrieve noisy KG information. Recent approaches leverage Graph Neural Networks (GNNs) to generate KG-based input embedding prefixes as soft prompts for LLMs but fail to account for question relevance, resulting in noisy prompts. Moreover, in MCQA tasks, the absence of relevant KG knowledge for certain answer options remains a significant challenge. To address these issues, we propose Question-Aware Knowledge Graph Prompting (QAP), which incorporates question embeddings into GNN aggregation to dynamically assess KG relevance. QAP employs global attention to capture inter-option relationships, enriching soft prompts with inferred knowledge. Experimental results demonstrate that QAP outperforms state-of-the-art methods across multiple datasets, highlighting its effectiveness.

arXiv:2503.23483

Order Independence With Finetuning

Large language models (LLMs) demonstrate remarkable performance on many NLP tasks, yet often exhibit order dependence: simply reordering semantically identical tokens (e.g., answer choices in multiple-choice questions) can lead to inconsistent predictions. Recent work proposes Set-Based Prompting (SBP) as a way to remove order information from designated token subsets, thereby mitigating positional biases. However, applying SBP on base models induces an out-of-distribution input format, which can degrade in-distribution performance. We introduce a fine-tuning strategy that integrates SBP into the training process, "pulling" these set-formatted prompts closer to the model's training manifold. We show that SBP can be incorporated into a model via fine-tuning. Our experiments on in-distribution (MMLU) and out-of-distribution (CSQA, ARC Challenge) multiple-choice tasks show that SBP fine-tuning significantly improves accuracy and robustness to answer-order permutations, all while preserving broader language modeling capabilities. We discuss the broader implications of order-invariant modeling and outline future directions for building fairer, more consistent LLMs.

arXiv:2503.24378

ACPBench Hard: Unrestrained Reasoning about Action, Change, and Planning

The ACPBench dataset provides atomic reasoning tasks required for efficient planning. The dataset is aimed at distilling the complex plan generation task into separate atomic reasoning tasks in their easiest possible form, boolean or multiple-choice questions, where the model has to choose the right answer from the provided options. While the aim of ACPBench is to test the simplest form of reasoning about action and change, when tasked with planning, a model does not typically have options to choose from and thus the reasoning required for planning dictates an open-ended, generative form for these tasks. To that end, we introduce ACPBench Hard, a generative version of ACPBench, with open-ended questions which the model needs to answer. Models that perform well on these tasks could in principle be integrated into a planner or be used directly as a policy. We discuss the complexity of these tasks as well as the complexity of validating the correctness of their answers and present validation algorithms for each task. Equipped with these validators, we test the performance of a variety of models on our tasks and find that for most of these tasks the performance of even the largest models is still subpar. Our experiments show that no model outperforms another in these tasks and with a few exceptions all tested language models score below 65%, indicating that even the current frontier language models have a long way to go before they can reliably reason about planning. In fact, even the so-called reasoning models struggle with solving these reasoning tasks. ACPBench Hard collection is available at the following link: https://ibm.github.io/ACPBench

arXiv:2503.24376공식 코드 · Papers with Code

Exploring the Effect of Reinforcement Learning on Video Understanding: Insights from SEED-Bench-R1

Recent advancements in Chain of Thought (COT) generation have significantly improved the reasoning capabilities of Large Language Models (LLMs), with reinforcement learning (RL) emerging as an effective post-training approach. Multimodal Large Language Models (MLLMs) inherit this reasoning potential but remain underexplored in tasks requiring both perception and logical reasoning. To address this, we introduce SEED-Bench-R1, a benchmark designed to systematically evaluate post-training methods for MLLMs in video understanding. It includes intricate real-world videos and complex everyday planning tasks in the format of multiple-choice questions, requiring sophisticated perception and reasoning. SEED-Bench-R1 assesses generalization through a three-level hierarchy: in-distribution, cross-environment, and cross-environment-task scenarios, equipped with a large-scale training dataset with easily verifiable ground-truth answers. Using Qwen2-VL-Instruct-7B as a base model, we compare RL with supervised fine-tuning (SFT), demonstrating RL's data efficiency and superior performance on both in-distribution and out-of-distribution tasks, even outperforming SFT on general video understanding benchmarks like LongVideoBench. Our detailed analysis reveals that RL enhances visual perception but often produces less logically coherent reasoning chains. We identify key limitations such as inconsistent reasoning and overlooked visual cues, and suggest future improvements in base model reasoning, reward modeling, and RL robustness against noisy signals.

arXiv:2503.19168공식 코드 · Papers with Code

Language Model Uncertainty Quantification with Attention Chain

Accurately quantifying a large language model's (LLM) predictive uncertainty is crucial for judging the reliability of its answers. While most existing research focuses on short, directly answerable questions with closed-form outputs (e.g., multiple-choice), involving intermediate reasoning steps in LLM responses is increasingly important. This added complexity complicates uncertainty quantification (UQ) because the probabilities assigned to answer tokens are conditioned on a vast space of preceding reasoning tokens. Direct marginalization is infeasible, and the dependency inflates probability estimates, causing overconfidence in UQ. To address this, we propose UQAC, an efficient method that narrows the reasoning space to a tractable size for marginalization. UQAC iteratively constructs an "attention chain" of tokens deemed "semantically crucial" to the final answer via a backtracking procedure. Starting from the answer tokens, it uses attention weights to identify the most influential predecessors, then iterates this process until reaching the input tokens. Similarity filtering and probability thresholding further refine the resulting chain, allowing us to approximate the marginal probabilities of the answer tokens, which serve as the LLM's confidence. We validate UQAC on multiple reasoning benchmarks with advanced open-source LLMs, demonstrating that it consistently delivers reliable UQ estimates with high computational efficiency.

arXiv:2503.20786공식 코드 · Papers with Code

Mobile-MMLU: A Mobile Intelligence Language Understanding Benchmark

Rapid advancements in large language models (LLMs) have increased interest in deploying them on mobile devices for on-device AI applications. Mobile users interact differently with LLMs compared to desktop users, creating unique expectations and data biases. Current benchmark datasets primarily target at server and desktop environments, and there is a notable lack of extensive datasets specifically designed for mobile contexts. Additionally, mobile devices face strict limitations in storage and computing resources, constraining model size and capabilities, thus requiring optimized efficiency and prioritized knowledge. To address these challenges, we introduce Mobile-MMLU, a large-scale benchmark dataset tailored for mobile intelligence. It consists of 16,186 questions across 80 mobile-related fields, designed to evaluate LLM performance in realistic mobile scenarios. A challenging subset, Mobile-MMLU-Pro, provides advanced evaluation similar in size to MMLU-Pro but significantly more difficult than our standard full set. Both benchmarks use multiple-choice, order-invariant questions focused on practical mobile interactions, such as recipe suggestions, travel planning, and essential daily tasks. The dataset emphasizes critical mobile-specific metrics like inference latency, energy consumption, memory usage, and response quality, offering comprehensive insights into model performance under mobile constraints. Moreover, it prioritizes privacy and adaptability, assessing models' ability to perform on-device processing, maintain user privacy, and adapt to personalized usage patterns. Mobile-MMLU family offers a standardized framework for developing and comparing mobile-optimized LLMs, enabling advancements in productivity and decision-making within mobile computing environments. Our code and data are available at: https://github.com/VILA-Lab/Mobile-MMLU.

arXiv:2503.18172

Unmasking Deceptive Visuals: Benchmarking Multimodal Large Language Models on Misleading Chart Question Answering

Misleading chart visualizations, which intentionally manipulate data representations to support specific claims, can distort perceptions and lead to incorrect conclusions. Despite decades of research, misleading visualizations remain a widespread and pressing issue. Recent advances in multimodal large language models (MLLMs) have demonstrated strong chart comprehension capabilities, yet no existing work has systematically evaluated their ability to detect and interpret misleading charts. This paper introduces the Misleading Chart Question Answering (Misleading ChartQA) Benchmark, a large-scale multimodal dataset designed to assess MLLMs in identifying and reasoning about misleading charts. It contains over 3,000 curated examples, covering 21 types of misleaders and 10 chart types. Each example includes standardized chart code, CSV data, and multiple-choice questions with labeled explanations, validated through multi-round MLLM checks and exhausted expert human review. We benchmark 16 state-of-the-art MLLMs on our dataset, revealing their limitations in identifying visually deceptive practices. We also propose a novel pipeline that detects and localizes misleaders, enhancing MLLMs' accuracy in misleading chart interpretation. Our work establishes a foundation for advancing MLLM-driven misleading chart comprehension. We publicly release the sample dataset to support further research in this critical area.

arXiv:2503.17599

Evaluating Clinical Competencies of Large Language Models with a General Practice Benchmark

Large Language Models (LLMs) have demonstrated considerable potential in general practice. However, existing benchmarks and evaluation frameworks primarily depend on exam-style or simplified question-answer formats, lacking a competency-based structure aligned with the real-world clinical responsibilities encountered in general practice. Consequently, the extent to which LLMs can reliably fulfill the duties of general practitioners (GPs) remains uncertain. In this work, we propose a novel evaluation framework to assess the capability of LLMs to function as GPs. Based on this framework, we introduce a general practice benchmark (GPBench), whose data are meticulously annotated by domain experts in accordance with routine clinical practice standards. We evaluate ten state-of-the-art LLMs and analyze their competencies. Our findings indicate that current LLMs are not yet ready for deployment in such settings without human oversight, and further optimization specifically tailored to the daily responsibilities of GPs is essential.

arXiv:2503.17485

SaudiCulture: A Benchmark for Evaluating Large Language Models Cultural Competence within Saudi Arabia

Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language processing; however, they often struggle to accurately capture and reflect cultural nuances. This research addresses this challenge by focusing on Saudi Arabia, a country characterized by diverse dialects and rich cultural traditions. We introduce SaudiCulture, a novel benchmark designed to evaluate the cultural competence of LLMs within the distinct geographical and cultural contexts of Saudi Arabia. SaudiCulture is a comprehensive dataset of questions covering five major geographical regions, such as West, East, South, North, and Center, along with general questions applicable across all regions. The dataset encompasses a broad spectrum of cultural domains, including food, clothing, entertainment, celebrations, and crafts. To ensure a rigorous evaluation, SaudiCulture includes questions of varying complexity, such as open-ended, single-choice, and multiple-choice formats, with some requiring multiple correct answers. Additionally, the dataset distinguishes between common cultural knowledge and specialized regional aspects. We conduct extensive evaluations on five LLMs, such as GPT-4, Llama 3.3, FANAR, Jais, and AceGPT, analyzing their performance across different question types and cultural contexts. Our findings reveal that all models experience significant performance declines when faced with highly specialized or region-specific questions, particularly those requiring multiple correct responses. Additionally, certain cultural categories are more easily identifiable than others, further highlighting inconsistencies in LLMs cultural understanding. These results emphasize the importance of incorporating region-specific knowledge into LLMs training to enhance their cultural competence.

arXiv:2503.16036공식 코드 · Papers with Code

Hybrid-Level Instruction Injection for Video Token Compression in Multi-modal Large Language Models

Recent Multi-modal Large Language Models (MLLMs) have been challenged by the computational overhead resulting from massive video frames, often alleviated through compression strategies. However, the visual content is not equally contributed to user instructions, existing strategies (\eg, average pool) inevitably lead to the loss of potentially useful information. To tackle this, we propose the Hybrid-level Instruction Injection Strategy for Conditional Token Compression in MLLMs (HICom), utilizing the instruction as a condition to guide the compression from both local and global levels. This encourages the compression to retain the maximum amount of user-focused information while reducing visual tokens to minimize computational burden. Specifically, the instruction condition is injected into the grouped visual tokens at the local level and the learnable tokens at the global level, and we conduct the attention mechanism to complete the conditional compression. From the hybrid-level compression, the instruction-relevant visual parts are highlighted while the temporal-spatial structure is also preserved for easier understanding of LLMs. To further unleash the potential of HICom, we introduce a new conditional pre-training stage with our proposed dataset HICom-248K. Experiments show that our HICom can obtain distinguished video understanding ability with fewer tokens, increasing the performance by 2.43\% average on three multiple-choice QA benchmarks and saving 78.8\% tokens compared with the SOTA method. The code is available at https://github.com/lntzm/HICom.

arXiv:2503.15837공식 코드 · Papers with Code

Fùxì: A Benchmark for Evaluating Language Models on Ancient Chinese Text Understanding and Generation

Ancient Chinese text processing presents unique challenges for large language models (LLMs) due to its distinct linguistic features, complex structural constraints, and rich cultural context. While existing benchmarks have primarily focused on evaluating comprehension through multiple-choice questions, there remains a critical gap in assessing models' generative capabilities in classical Chinese. We introduce F\`ux\`i, a comprehensive benchmark that evaluates both understanding and generation capabilities across 21 diverse tasks. Our benchmark distinguishes itself through three key contributions: (1) balanced coverage of both comprehension and generation tasks, including novel tasks like poetry composition and couplet completion, (2) specialized evaluation metrics designed specifically for classical Chinese text generation, combining rule-based verification with fine-tuned LLM evaluators, and (3) a systematic assessment framework that considers both linguistic accuracy and cultural authenticity. Through extensive evaluation of state-of-the-art LLMs, we reveal significant performance gaps between understanding and generation tasks, with models achieving promising results in comprehension but struggling considerably in generation tasks, particularly those requiring deep cultural knowledge and adherence to classical formats. Our findings highlight the current limitations in ancient Chinese text processing and provide insights for future model development. The benchmark, evaluation toolkit, and baseline results are publicly available to facilitate research in this domain.

arXiv:2503.15525

The Use of Artificial Intelligence Tools in Assessing Content Validity: A Comparative Study with Human Experts

In this study, it was investigated whether AI evaluators assess the content validity of B1-level English reading comprehension test items in a manner similar to human evaluators. A 25-item multiple-choice test was developed, and these test items were evaluated by four human and four AI evaluators. No statistically significant difference was found between the scores given by human and AI evaluators, with similar evaluation trends observed. The Content Validity Ratio (CVR) and the Item Content Validity Index (I-CVI) were calculated and analyzed using the Wilcoxon Signed-Rank Test, with no statistically significant difference. The findings revealed that in some cases, AI evaluators could replace human evaluators. However, differences in specific items were thought to arise from varying interpretations of the evaluation criteria. Ensuring linguistic clarity and clearly defining criteria could contribute to more consistent evaluations. In this regard, the development of hybrid evaluation systems, in which AI technologies are used alongside human experts, is recommended.

arXiv:2503.16167

CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models

State-of-the-art large language models (LLMs) have demonstrated impressive code generation capabilities but struggle with real-world software engineering tasks, such as revising source code to address code reviews, hindering their practical use. Code review comments are often implicit, ambiguous, and colloquial, requiring models to grasp both code and human intent. This challenge calls for evaluating large language models' ability to bridge both technical and conversational contexts. While existing work has employed the automated code refinement (ACR) task to resolve these comments, current evaluation methods fall short, relying on text matching metrics that provide limited insight into model failures and remain susceptible to training data contamination. To address these limitations, we introduce a novel evaluation benchmark, $\textbf{CodeReviewQA}$ that enables us to conduct fine-grained assessment of model capabilities and mitigate data contamination risks. In CodeReviewQA, we decompose the generation task of code refinement into $\textbf{three essential reasoning steps}$: $\textit{change type recognition}$ (CTR), $\textit{change localisation}$ (CL), and $\textit{solution identification}$ (SI). Each step is reformulated as multiple-choice questions with varied difficulty levels, enabling precise assessment of model capabilities, while mitigating data contamination risks. Our comprehensive evaluation spans 72 recently released large language models on $\textbf{900 manually curated, high-quality examples}$ across nine programming languages. Our results show that CodeReviewQA is able to expose specific model weaknesses in code review comprehension, disentangled from their generative automated code refinement results.

arXiv:2503.15778

AutoDrive-QA- Automated Generation of Multiple-Choice Questions for Autonomous Driving Datasets Using Large Vision-Language Models

In autonomous driving, open-ended question answering often suffers from unreliable evaluations because freeform responses require either complex metrics or subjective human judgment. To address this challenge, we introduce AutoDrive-QA, an automatic pipeline that converts existing driving QA datasets (including DriveLM, NuScenes-QA, and LingoQA) into a structured multiple-choice question (MCQ) format. This benchmark systematically assesses perception, prediction, and planning tasks, providing a standardized and objective evaluation framework. AutoDrive-QA employs an automated pipeline that leverages large language models (LLMs) to generate high-quality, contextually relevant distractors based on domain-specific error patterns commonly found in autonomous driving scenarios. To evaluate both general capabilities and generalization performance, we test the benchmark on three public datasets and conduct zero-shot experiments on an unseen dataset. The zero-shot evaluations reveal that GPT-4V leads with 69.57% accuracy -- achieving 74.94% in Perception, 65.33% in Prediction, and 68.45% in Planning -- demonstrating that while all models excel in Perception, they struggle in Prediction. Consequently, AutoDrive-QA establishes a rigorous, unbiased standard for integrating and evaluating different vision-language models across various autonomous driving datasets, thereby improving generalization in this field. We release all the codes in the AutoDrive-QA GitHub Repository.

arXiv:2503.14391공식 코드 · Papers with Code

How much do LLMs learn from negative examples?

Large language models (LLMs) undergo a three-phase training process: unsupervised pre-training, supervised fine-tuning (SFT), and learning from human feedback (RLHF/DPO). Notably, it is during the final phase that these models are exposed to negative examples -- incorrect, rejected, or suboptimal responses to queries. This paper delves into the role of negative examples in the training of LLMs, using a likelihood-ratio (Likra) model on multiple-choice question answering benchmarks to precisely manage the influence and the volume of negative examples. Our findings reveal three key insights: (1) During a critical phase in training, Likra with negative examples demonstrates a significantly larger improvement per training example compared to SFT using only positive examples. This leads to a sharp jump in the learning curve for Likra unlike the smooth and gradual improvement of SFT; (2) negative examples that are plausible but incorrect (near-misses) exert a greater influence; and (3) while training with positive examples fails to significantly decrease the likelihood of plausible but incorrect answers, training with negative examples more accurately identifies them. These results indicate a potentially significant role for negative examples in improving accuracy and reducing hallucinations for LLMs.