Daily Papers

Large language models (LLMs) have demonstrated remarkable progress in understanding long-context inputs. However, benchmarks for evaluating the long-context reasoning abilities of LLMs fall behind the pace. Existing benchmarks often focus on a narrow range of tasks or those that do not demand complex reasoning. To address this gap and enable a more comprehensive evaluation of the long-context reasoning capabilities of current LLMs, we propose a new synthetic benchmark, LongReason, which is constructed by synthesizing long-context reasoning questions from a varied set of short-context reasoning questions through context expansion. LongReason consists of 794 multiple-choice reasoning questions with diverse reasoning patterns across three task categories: reading comprehension, logical inference, and mathematical word problems. We evaluate 21 LLMs on LongReason, revealing that most models experience significant performance drops as context length increases. Our further analysis shows that even state-of-the-art LLMs still have significant room for improvement in providing robust reasoning across different tasks. We will open-source LongReason to support the comprehensive evaluation of LLMs' long-context reasoning capabilities.

We introduce HEAD-QA v2, an expanded and updated version of a Spanish/English healthcare multiple-choice reasoning dataset originally released by Vilares and Gómez-Rodríguez (2019). The update responds to the growing need for high-quality datasets that capture the linguistic and conceptual complexity of healthcare reasoning. We extend the dataset to over 12,000 questions from ten years of Spanish professional exams, benchmark several open-source LLMs using prompting, RAG, and probability-based answer selection, and provide additional multilingual versions to support future work. Results indicate that performance is mainly driven by model scale and intrinsic reasoning ability, with complex inference strategies obtaining limited gains. Together, these results establish HEAD-QA v2 as a reliable resource for advancing research on biomedical reasoning and model improvement.

We introduce SEA-HELM Audio, a Southeast Asian (SEA) multilingual evaluation framework that extends the text-based SEA-HELM benchmark to assess performance degradation in cascaded automatic speech recognition and large language model (ASR-LLM) pipelines. Using synthesized speech, we compare speech-input and text-input performance across five languages (Chinese, English, Malay, Indonesian, and Tamil) on multiple-choice reasoning and instruction-following tasks. We evaluate open-weight and commercial ASR-LLM systems using identical verifiers and quantify the speech gap as the difference between text and speech accuracy. Results show substantial degradation, with accuracy drops of 9.00 to 18.50 percentage points for multiple-choice reasoning and 20.00 to 42.86 points for instruction following. Tamil exhibits the largest degradation, exceeding 40 percentage points. Our findings highlight the limitations of text-only benchmarks in predicting voice-agent reliability and demonstrate the value of SEA-HELM Audio for evaluating multilingual voice-agent systems.

Brazilian Portuguese remains under-served by open language models, and the few that exist are difficult to reproduce and are often compared without measures of uncertainty. We release Manacá-1B, an open decoder-only model of 1.72 billion parameters trained from scratch for Brazilian Portuguese with a fully containerized, reproducible pipeline. The pretraining is stable, with zero skipped or NaN steps and self-recovering loss spikes, and we release its full log and dynamics. We evaluate the model against nine open baselines on four Portuguese benchmarks under a single harness. Every comparison reports a standard error and a paired significance test, and the harness is validated against previously published numbers. On last-word prediction Manacá-1B is the strongest model below the 7B scale, exceeding both Tucano-1b1 and Tucano-2b4 on LAMBADA-PT with large paired margins; it is competitive on commonsense completion and near chance on multiple-choice reasoning, as are all small base models. Along the way we document a concrete evaluation pitfall: converting a SentencePiece tokenizer with case-folding normalization to the HuggingFace fast format silently drops the normalizer, routing every capitalized token to byte-fallback and depressing scores in a way that is invisible in aggregate metrics. The uncorrected tokenizer lowered LAMBADA-PT accuracy from 45.3 to 25.0; we quantify the effect and provide a one-line fix that reproduces the training tokenizer exactly. Code, raw training and evaluation logs, per-example prediction vectors, the model weights, and the corrected tokenizer are released so that every number in this paper can be recomputed.

One of the most widely used methods to evaluate LLMs are Multiple Choice Question (MCQ) tests. MCQ benchmarks enable the testing of LLM knowledge on almost any topic at scale as the results can be processed automatically. To help the LLM answer, a few examples called few shots can be included in the prompt. Moreover, the LLM can be asked to answer the question directly with the selected option or to first provide the reasoning and then the selected answer, which is known as chain of thought. In addition to checking whether the selected answer is correct, the evaluation can look at the LLM-estimated probability of its response as an indication of the confidence of the LLM in the response. In this paper, we study how the LLM confidence in its answer depends on whether the model has been asked to answer directly or to provide the reasoning before answering. The results of the evaluation of questions on a wide range of topics in seven different models show that LLMs are more confident in their answers when they provide reasoning before the answer. This occurs regardless of whether the selected answer is correct. Our hypothesis is that this behavior is due to the reasoning that modifies the probability of the selected answer, as the LLM predicts the answer based on the input question and the reasoning that supports the selection made. Therefore, LLM estimated probabilities seem to have intrinsic limitations that should be understood in order to use them in evaluation procedures. Interestingly, the same behavior has been observed in humans, for whom explaining an answer increases confidence in its correctness.

Prompt engineering and calibration make large language models excel at reasoning tasks, including multiple choice commonsense reasoning. From a practical perspective, we investigate and evaluate these strategies on smaller language models. Through experiments on five commonsense reasoning benchmarks, we find that each strategy favors certain models, but their joint effects are mostly negative.

There is a significant gap in evaluating cultural reasoning in LLMs using conversational datasets that capture culturally rich and dialectal contexts. Most Arabic benchmarks focus on short text snippets in Modern Standard Arabic (MSA), overlooking the cultural nuances that naturally arise in dialogues. To address this gap, we introduce ArabCulture-Dialogue, a culturally grounded conversational dataset covering 13 Arabic-speaking countries, in both MSA and each country's respective dialect, spanning 12 daily-life topics and 54 fine-grained subtopics. We utilize the dataset to form three benchmarking tasks: (i) multiple-choice cultural reasoning, (ii) machine translation between MSA and dialects, and (iii) dialect-steering generation. Our experiments indicate that the performance gap between MSA and Arabic dialects still exists, whereby the models perform worse on all three tasks in the dialectal setup, compared to the MSA one.

Transformer-based large language models (LLMs) use the key-value (KV) cache to significantly accelerate inference by storing the key and value embeddings of past tokens. However, this cache consumes significant GPU memory. In this work, we introduce HashEvict, an algorithm that uses locality-sensitive hashing (LSH) to compress the KV cache. HashEvict quickly locates tokens in the cache that are cosine dissimilar to the current query token. This is achieved by computing the Hamming distance between binarized Gaussian projections of the current token query and cached token keys, with a projection length much smaller than the embedding dimension. We maintain a lightweight binary structure in GPU memory to facilitate these calculations. Unlike existing compression strategies that compute attention to determine token retention, HashEvict makes these decisions pre-attention, thereby reducing computational costs. Additionally, HashEvict is dynamic - at every decoding step, the key and value of the current token replace the embeddings of a token expected to produce the lowest attention score. We demonstrate that HashEvict can compress the KV cache by 30%-70% while maintaining high performance across reasoning, multiple-choice, long-context retrieval and summarization tasks.

Evaluating the symbolic reasoning of large language models (LLMs) calls for geometry benchmarks that require multi-step proofs grounded in both text and diagrams. However, existing benchmarks are often limited in scale and rarely provide visually grounded multiple-choice questions, limiting reliable evaluation of complex reasoning. We introduce GeoChallenge, a dataset of 90K automatically generated multiple-choice geometry proof problems, each requiring multi-step reasoning over aligned textual descriptions and diagrams. GeoChallenge provides fine-grained complexity ratings and formal language annotations to enable controlled evaluation. Experiments on multiple advanced LLMs show a clear performance gap between models and humans (the best-performing model, GPT-5-nano, achieves 75.89 exact match vs. 94.74 for humans). Further analysis also reveals three common failure patterns of LLMs: (1) exact match failures under the multiple-choice setting; (2) weak visual reliance; and (3) overextended reasoning without convergence.

Reasoning quality in large language models depends not only on producing correct answers but also on generating valid intermediate steps. We study this through multiple-choice question answering (MCQA), which provides a controlled setting with fixed answer options. Our analysis shows that when questions are effectively unsolvable for a model, spurious chains of thought (CoTs) are more likely to appear, leading to false positives. By estimating the solvability of each question, we uncover an intermediate regime where learning is most effective. Building on this insight, we adapt outcome-supervised reward models and reinforcement learning with group-relative advantage to incorporate solvability into their objectives. Across experiments on math and multimodal datasets, these modifications consistently yield higher rates of process-correct reasoning and, in reinforcement learning, improved answer accuracy as well. Our results highlight solvability as a key factor for reducing hallucinations and increasing reliability in CoT reasoning.

Multimodal Large Language Models (MLLMs) have demonstrated remarkable multimodal understanding capabilities in Visual Question Answering (VQA) tasks by integrating visual and textual features. However, under the challenging ten-choice question evaluation paradigm, existing methods still exhibit significant limitations when processing PDF documents with complex layouts and lengthy content. Notably, current mainstream models suffer from a strong bias toward English training data, resulting in suboptimal performance for Japanese and other language scenarios. To address these challenges, this paper proposes a novel Japanese PDF document understanding framework that combines multimodal hierarchical reasoning mechanisms with Colqwen-optimized retrieval methods, while innovatively introducing a semantic verification strategy through sub-question decomposition. Experimental results demonstrate that our framework not only significantly enhances the model's deep semantic parsing capability for complex documents, but also exhibits superior robustness in practical application scenarios.

Large Language Models (LLMs) like ChatGPT demonstrate significant potential in the medical field, often evaluated using multiple-choice questions (MCQs) similar to those found on the USMLE. Despite their prevalence in medical education, MCQs have limitations that might be exacerbated when assessing LLMs. To evaluate the effectiveness of MCQs in assessing the performance of LLMs, we developed a fictional medical benchmark focused on a non-existent gland, the Glianorex. This approach allowed us to isolate the knowledge of the LLM from its test-taking abilities. We used GPT-4 to generate a comprehensive textbook on the Glianorex in both English and French and developed corresponding multiple-choice questions in both languages. We evaluated various open-source, proprietary, and domain-specific LLMs using these questions in a zero-shot setting. The models achieved average scores around 67%, with minor performance differences between larger and smaller models. Performance was slightly higher in English than in French. Fine-tuned medical models showed some improvement over their base versions in English but not in French. The uniformly high performance across models suggests that traditional MCQ-based benchmarks may not accurately measure LLMs' clinical knowledge and reasoning abilities, instead highlighting their pattern recognition skills. This study underscores the need for more robust evaluation methods to better assess the true capabilities of LLMs in medical contexts.

We present GSM-MC, a multiple-choice (MC) dataset constructed by collecting answers and incorrect predictions on GSM8K from 60 open-source models. Through extensive experiments, we show that LLMs' performance on the MC version of this popular benchmark is strongly correlated with their performance on the original version and is quite robust to distractor choices and option orders, while the evaluation time is reduced by a factor of up to 30. Following similar procedures, we introduce MATH-MC, constructed from MATH, and PythonIO, a new program reasoning MC dataset constructed from HumanEval and MBPP. Experimental results indicate that LLMs' performance on these MC benchmarks leaves much room for improvement. Our data and code are available at https://github.com/Geralt-Targaryen/MC-Evaluation.

Long-context multiple-choice question answering tasks require robust reasoning over extensive text sources. Since most of the pre-trained transformer models are restricted to processing only a few hundred words at a time, successful completion of such tasks often relies on the identification of evidence spans, such as sentences, that provide supporting evidence for selecting the correct answer. Prior research in this domain has predominantly utilized pre-trained dense retrieval models, given the absence of supervision to fine-tune the retrieval process. This paper proposes a novel method called Options Aware Dense Retrieval (OADR) to address these challenges. ORDA uses an innovative approach to fine-tuning retrieval by leveraging query-options embeddings, which aim to mimic the embeddings of the oracle query (i.e., the query paired with the correct answer) for enhanced identification of supporting evidence. Through experiments conducted on the QuALITY benchmark dataset, we demonstrate that our proposed model surpasses existing baselines in terms of performance and accuracy.

Large language models (LLMs) now give reasoning before answering, excelling in tasks like multiple-choice question answering (MCQA). Yet, a concern is that LLMs do not solve MCQs as intended, as work finds LLMs sans reasoning succeed in MCQA without using the question, i.e., choices-only. Such partial-input success is often linked to trivial shortcuts, but reasoning traces could reveal if choices-only strategies are truly shallow. To examine these strategies, we have reasoning LLMs solve MCQs in full and choices-only inputs; test-time reasoning often boosts accuracy in full and in choices-only, half the time. While possibly due to shallow shortcuts, choices-only success is barely affected by the length of reasoning traces, and after finding traces pass faithfulness tests, we show they use less problematic strategies like inferring missing questions. In all, we challenge claims that partial-input success is always a flaw, so we propose how reasoning traces could separate problematic data from less problematic reasoning.

Medical multiple-choice question answering (MCQA) is particularly difficult. Questions may describe patient symptoms and ask for the correct diagnosis, which requires domain knowledge and complex reasoning. Standard language modeling pretraining alone is not sufficient to achieve the best results. jin2020disease showed that focusing masked language modeling on disease name prediction when using medical encyclopedic paragraphs as input leads to considerable MCQA accuracy improvement. In this work, we show that (1) fine-tuning on generated MCQA dataset outperforms the masked language modeling based objective and (2) correctly masking the cues to the answers is critical for good performance. We release new pretraining datasets and achieve state-of-the-art results on 4 MCQA datasets, notably +5.7\% with base-size model on MedQA-USMLE.

While visual question-answering (VQA) benchmarks have catalyzed the development of reasoning techniques, they have focused on vertical thinking. Effective problem-solving also necessitates lateral thinking, which remains understudied in AI and has not been used to test visual perception systems. To bridge this gap, we formulate visual lateral thinking as a multiple-choice question-answering task and describe a three-step taxonomy-driven methodology for instantiating task examples. Then, we develop COLUMBUS, a synthetic benchmark that applies the task pipeline to create QA sets with text and icon rebus puzzles based on publicly available collections of compounds and common phrases. COLUMBUS comprises over 1,000 puzzles, each with four answer candidates. While the SotA vision-language models (VLMs) achieve decent performance, our evaluation demonstrates a substantial gap between humans and models. VLMs benefit from human-curated descriptions but struggle to self-generate such representations at the right level of abstraction.

We introduce WiCkeD, a simple method to increase the complexity of existing multiple-choice benchmarks by randomly replacing a choice with "None of the above", a method often used in educational tests. We show that WiCkeD can be automatically applied to any existing benchmark, making it more challenging. We apply WiCkeD to 6 popular benchmarks and use it to evaluate 18 open-weight LLMs. The performance of the models drops 12.1 points on average with respect to the original versions of the datasets. When using chain-of-thought on 3 MMLU datasets, the performance drop for the WiCkeD variant is similar to the one observed when using the LLMs directly, showing that WiCkeD is also challenging for models with enhanced reasoning abilities. WiCkeD also uncovers that some models are more sensitive to the extra reasoning required, providing additional information with respect to the original benchmarks. We relase our code and data at https://github.com/ahmedselhady/wicked-benchmarks.

The rapid development of large language models (LLMs) has shown promising practical results. However, their low interpretability often leads to errors in unforeseen circumstances, limiting their utility. Many works have focused on creating comprehensive evaluation systems, but previous benchmarks have primarily assessed problem-solving abilities while neglecting the response's uncertainty, which may result in unreliability. Recent methods for measuring LLM reliability are resource-intensive and unable to test black-box models. To address this, we propose UBENCH, a comprehensive benchmark for evaluating LLM reliability. UBENCH includes 3,978 multiple-choice questions covering knowledge, language, understanding, and reasoning abilities. Experimental results show that UBENCH has achieved state-of-the-art performance, while its single-sampling method significantly saves computational resources compared to baseline methods that require multiple samplings. Additionally, based on UBENCH, we evaluate the reliability of 15 popular LLMs, finding GLM4 to be the most outstanding, closely followed by GPT-4. We also explore the impact of Chain-of-Thought prompts, role-playing prompts, option order, and temperature on LLM reliability, analyzing the varying effects on different LLMs.

Large Language Models (LLMs) have shown remarkable capabilities, not only in generating human-like text, but also in acquiring knowledge. This highlights the need to go beyond the typical Natural Language Processing downstream benchmarks and asses the various aspects of LLMs including knowledge and reasoning. Numerous benchmarks have been developed to evaluate LLMs knowledge, but they predominantly focus on the English language. Given that many LLMs are multilingual, relying solely on benchmarking English knowledge is insufficient. To address this issue, we introduce AraSTEM, a new Arabic multiple-choice question dataset aimed at evaluating LLMs knowledge in STEM subjects. The dataset spans a range of topics at different levels which requires models to demonstrate a deep understanding of scientific Arabic in order to achieve high accuracy. Our findings show that publicly available models of varying sizes struggle with this dataset, and underscores the need for more localized language models. The dataset is freely accessible on Hugging Face.

We analyzed effectiveness of three generative pre-trained transformer (GPT) models in answering multiple-choice question (MCQ) assessments, often involving short snippets of code, from introductory and intermediate programming courses at the postsecondary level. This emerging technology stirs countless discussions of its potential uses (e.g., exercise generation, code explanation) as well as misuses in programming education (e.g., cheating). However, the capabilities of GPT models and their limitations to reason about and/or analyze code in educational settings have been under-explored. We evaluated several OpenAI's GPT models on formative and summative MCQ assessments from three Python courses (530 questions). We found that MCQs containing code snippets are not answered as successfully as those that only contain natural language. While questions requiring to fill-in a blank in the code or completing a natural language statement about the snippet are handled rather successfully, MCQs that require analysis and/or reasoning about the code (e.g., what is true/false about the snippet, or what is its output) appear to be the most challenging. These findings can be leveraged by educators to adapt their instructional practices and assessments in programming courses, so that GPT becomes a valuable assistant for a learner as opposed to a source of confusion and/or potential hindrance in the learning process.

With the advancement of large language models (LLMs), their performance on multiple-choice question (MCQ) tasks has improved significantly. However, existing approaches face key limitations: answer choices are typically presented to LLMs without contextual grounding or explanation. This absence of context can lead to incomplete exploration of all possible answers, ultimately degrading the models' reasoning capabilities. To address these challenges, we introduce BiasPrompting, a novel inference framework that guides LLMs to generate and critically evaluate reasoning across all plausible answer options before reaching a final prediction. It consists of two components: first, a reasoning generation stage, where the model is prompted to produce supportive reasonings for each answer option, and then, a reasoning-guided agreement stage, where the generated reasonings are synthesized to select the most plausible answer. Through comprehensive evaluations, BiasPrompting demonstrates significant improvements in five widely used multiple-choice question answering benchmarks. Our experiments showcase that BiasPrompting enhances the reasoning capabilities of LLMs and provides a strong foundation for tackling complex and challenging questions, particularly in settings where existing methods underperform.

Scientific figure multiple-choice question answering (MCQA) requires models to reason over diverse visual evidence, ranging from charts and multipanel figures to microscopy and biomedical images. However, this setting suffers from a distinctive bias: answer choices themselves can act as priors, steering multimodal models toward scientifically plausible options even when the figure supports a different answer. We investigate this failure mode through a simple question: what if decoding explicitly discounts what the model would prefer from text alone, so as to favor figure-grounded evidence? To this end, we propose SCICON, a training-free decoding method that scores each candidate by subtracting a text-only option score from its image-conditioned counterpart. Unlike prior contrastive decoding approaches that mitigate hallucinations by contrasting original inputs with distorted images or perturbed instructions, SCICON directly targets the choice-induced prior encoded in candidate text. Across three scientific figure QA benchmarks and three model backbones, SCICON consistently improves accuracy over standard decoding baselines. These results show that decoding against choice-induced priors is an effective and simple way to improve figure-grounded reasoning in scientific MCQA.

Large language models (LLMs) are increasingly evaluated on single-answer multiple-choice tasks, yet many real-world problems require identifying all correct answers from a set of options. This capability remains underexplored. We introduce SATA-BENCH, the first dedicated benchmark for evaluating LLMs on Select All That Apply (SATA) questions across diverse domains, including reading comprehension, law, and biomedicine. Our evaluation of 27 open-source and proprietary models reveals a significant gap: even the strongest model achieves only 41.8% exact match, exposing LLMs' inability to reliably identify all correct answers. We find that this weakness stems from two core challenges: selection bias - models favor certain choices regardless of content, and count bias - models fail to predict the correct number of answers. To address these issues, we propose Choice Funnel, a decoding strategy that combines token debiasing with adaptive thresholding to guide models toward complete and accurate selections. Choice Funnel achieves up to 29% higher exact match than competitive baselines while reducing inference cost by over 64%. Our findings expose fundamental limitations in current LLMs and introduce a new framework for diagnosing and improving multi-answer reasoning. We release SATA-BENCH and Choice Funnel to promote LLM development for robust decision-making in realistic, multi-answer applications.

We propose a new paradigm for zero-shot learners that is format agnostic, i.e., it is compatible with any format and applicable to a list of language tasks, such as text classification, commonsense reasoning, coreference resolution, and sentiment analysis. Zero-shot learning aims to train a model on a given task such that it can address new learning tasks without any additional training. Our approach converts zero-shot learning into multiple-choice tasks, avoiding problems in commonly used large-scale generative models such as FLAN. It not only adds generalization ability to models but also significantly reduces the number of parameters. Our method shares the merits of efficient training and deployment. Our approach shows state-of-the-art performance on several benchmarks and produces satisfactory results on tasks such as natural language inference and text classification. Our model achieves this success with only 235M parameters, which is substantially smaller than state-of-the-art models with billions of parameters. The code and pre-trained models are available at https://github.com/IDEA-CCNL/Fengshenbang-LM .

This paper presents FormationEval, an open multiple-choice question benchmark for evaluating language models on petroleum geoscience and subsurface disciplines. The dataset contains 505 questions across seven domains including petrophysics, petroleum geology and reservoir engineering, derived from three authoritative sources using a reasoning model with detailed instructions and a concept-based approach that avoids verbatim copying of copyrighted text. Each question includes source metadata to support traceability and audit. The evaluation covers 72 models from major providers including OpenAI, Anthropic, Google, Meta and open-weight alternatives. The top performers achieve over 97\% accuracy, with Gemini 3 Pro Preview reaching 99.8\%, while tier and domain gaps persist. Among open-weight models, GLM-4.7 leads at 98.6\%, with several DeepSeek, Llama, Qwen and Mistral models also exceeding 93\%. The performance gap between open-weight and closed models is narrower than expected, with several lower-cost open-weight models exceeding 90\% accuracy. Petrophysics emerges as the most challenging domain across all models, while smaller models show wider performance variance. Residual length bias in the dataset (correct answers tend to be longer) is documented along with bias mitigation strategies applied during construction. The benchmark, evaluation code and results are publicly available.

Large Language Models (LLMs) have demonstrated remarkable capabilities in various NLP tasks. However, previous works have shown these models are sensitive towards prompt wording, and few-shot demonstrations and their order, posing challenges to fair assessment of these models. As these models become more powerful, it becomes imperative to understand and address these limitations. In this paper, we focus on LLMs robustness on the task of multiple-choice questions -- commonly adopted task to study reasoning and fact-retrieving capability of LLMs. Investigating the sensitivity of LLMs towards the order of options in multiple-choice questions, we demonstrate a considerable performance gap of approximately 13% to 75% in LLMs on different benchmarks, when answer options are reordered, even when using demonstrations in a few-shot setting. Through a detailed analysis, we conjecture that this sensitivity arises when LLMs are uncertain about the prediction between the top-2/3 choices, and specific options placements may favor certain prediction between those top choices depending on the question caused by positional bias. We also identify patterns in top-2 choices that amplify or mitigate the model's bias toward option placement. We found that for amplifying bias, the optimal strategy involves positioning the top two choices as the first and last options. Conversely, to mitigate bias, we recommend placing these choices among the adjacent options. To validate our conjecture, we conduct various experiments and adopt two approaches to calibrate LLMs' predictions, leading to up to 8 percentage points improvement across different models and benchmarks.

In recent years, there have been significant breakthroughs in the field of natural language processing, particularly with the development of large language models (LLMs). These LLMs have showcased remarkable capabilities on various benchmarks. In the healthcare field, the exact role LLMs and other future AI models will play remains unclear. There is a potential for these models in the future to be used as part of adaptive physician training, medical co-pilot applications, and digital patient interaction scenarios. The ability of AI models to participate in medical training and patient care will depend in part on their mastery of the knowledge content of specific medical fields. This study investigated the medical knowledge capability of LLMs, specifically in the context of internal medicine subspecialty multiple-choice test-taking ability. We compared the performance of several open-source LLMs (Koala 7B, Falcon 7B, Stable-Vicuna 13B, and Orca Mini 13B), to GPT-4 and Claude 2 on multiple-choice questions in the field of Nephrology. Nephrology was chosen as an example of a particularly conceptually complex subspecialty field within internal medicine. The study was conducted to evaluate the ability of LLM models to provide correct answers to nephSAP (Nephrology Self-Assessment Program) multiple-choice questions. The overall success of open-sourced LLMs in answering the 858 nephSAP multiple-choice questions correctly was 17.1% - 25.5%. In contrast, Claude 2 answered 54.4% of the questions correctly, whereas GPT-4 achieved a score of 73.3%. We show that current widely used open-sourced LLMs do poorly in their ability for zero-shot reasoning when compared to GPT-4 and Claude 2. The findings of this study potentially have significant implications for the future of subspecialty medical training and patient care.

This paper examines the role of cognitive biases in the decision-making processes of large language models (LLMs), challenging the conventional goal of eliminating all biases. We show that certain cognitive biases when properly balanced, can enhance decision-making efficiency through rational deviations and heuristic shortcuts. By introducing heuristic moderation and an abstention option, which allows LLMs to withhold responses when uncertain, we reduce error rates, improve decision accuracy, and optimize decision rates. Using the Balance Rigor and Utility (BRU) dataset, developed through expert collaboration, our findings demonstrate that targeted inspection of cognitive biases aligns LLM decisions more closely with human reasoning, enhancing reliability and suggesting strategies for future improvements. This approach offers a novel way to leverage cognitive biases to improve the practical utility of LLMs across various applications.

Large language models have recently demonstrated remarkable abilities to self-correct their responses through iterative refinement, often referred to as self-consistency or self-reflection. However, the dynamics of this self-correction mechanism may differ substantially depending on whether the model is tasked with open-ended text generation or with selecting the most appropriate response from multiple predefined options. In this paper, we conduct a systematic investigation of these two paradigms by comparing performance trends and error-correction behaviors across various natural language understanding and reasoning tasks, covering language models of different scales and families. Our experimental results reveal distinct patterns of improvement and failure modes: While open-ended generation often benefits from the flexibility of re-interpretation and compositional refinement, multiple-choice selection can leverage clearer solution boundaries but may be limited by the provided options. This contrast also reflects the dual demands faced by emerging agentic LLM applications: effective agents must not only generate and refine open-ended plans or explanations, but also make reliable discrete choices when operating within constrained action spaces. Our findings, therefore, highlight that the design of self-correction mechanisms should take into account the interaction between task structure and output space, with implications for both knowledge-intensive reasoning and decision-oriented applications of LLMs.

Recent advances in large-scale language models (LLMs) have made multi-agent architectures attractive for challenging reasoning tasks. However, many existing systems rely on stochastic routing or ad-hoc heuristics, making their behavior difficult to reproduce and their decision process hard to interpret. We propose ORCH, a deterministic coordination framework for discrete-choice reasoning that orchestrates heterogeneous LLMs. ORCH follows a ``many analyses, one decision'' paradigm: multiple base models independently produce structured analyses, and a dedicated merge agent outputs the final choice. The framework uses fixed rules for task decomposition and answer aggregation, keeping the pipeline predictable, reproducible, and training-free. Determinism here refers to fixed routing and aggregation rules under a fixed evaluation protocol, rather than strict bit-level reproducibility across deployments. To exploit model complementarity, we optionally introduce an EMA-guided router that updates agent selection using historical accuracy, latency, or cost; since it relies on answer-based feedback, it is mainly intended for benchmarking, controlled evaluation, or delayed-feedback settings. Experiments on MMLU, MMLU-Pro, and GSM8K show that ORCH consistently outperforms single-model baselines and a majority-vote ensemble. On MMLU-Pro, ORCH improves accuracy by over 10 points compared to the strongest baseline, and on GSM8K it yields gains exceeding 50 points; McNemar tests confirm statistical significance. The EMA router provides an additional 0.7--2.0 point accuracy boost, and ablations show that both multi-agent collaboration and routing contribute substantially. Overall, ORCH offers a practical path toward controllable, interpretable, and deployment-ready LLM-based agent systems for discrete-choice reasoning.

One of the most widely used tasks to evaluate Large Language Models (LLMs) is Multiple-Choice Question Answering (MCQA). While open-ended question answering tasks are more challenging to evaluate, MCQA tasks are, in principle, easier to assess, as the model's answer is thought to be simple to extract and is directly compared to a set of predefined choices. However, recent studies have started to question the reliability of MCQA evaluation, showing that multiple factors can significantly impact the reported performance of LLMs, especially when the model generates free-form text before selecting one of the answer choices. In this work, we shed light on the inconsistencies of MCQA evaluation strategies, which can lead to inaccurate and misleading model comparisons. We systematically analyze whether existing answer extraction methods are aligned with human judgment, and how they are influenced by answer constraints in the prompt across different domains. Our experiments demonstrate that traditional evaluation strategies often underestimate LLM capabilities, while LLM-based answer extractors are prone to systematic errors. Moreover, we reveal a fundamental trade-off between including format constraints in the prompt to simplify answer extraction and allowing models to generate free-form text to improve reasoning. Our findings call for standardized evaluation methodologies and highlight the need for more reliable and consistent MCQA evaluation practices.

Large Language Models (LLMs) with reasoning capabilities have achieved state-of-the-art performance on a wide range of tasks. Despite its empirical success, the tasks and model scales at which reasoning becomes effective, as well as its training and inference costs, remain underexplored. In this work, we rely on a synthetic data distillation framework to conduct a large-scale supervised study. We compare Instruction Fine-Tuning (IFT) and reasoning models of varying sizes, on a wide range of math-centric and general-purpose tasks, evaluating both multiple-choice and open-ended formats. Our analysis reveals that reasoning consistently improves model performance, often matching or surpassing significantly larger IFT systems. Notably, while IFT remains Pareto-optimal in training and inference costs, reasoning models become increasingly valuable as model size scales, overcoming IFT performance limits on reasoning-intensive and open-ended tasks.

Multiple-choice QA benchmarks usually evaluate small language models (SLMs) as direct answerers, but deployed language-model systems increasingly rely on external scaffolds such as tools, code, and repeated model calls. We introduce Code-Guided Reasoning (CGR), an evaluation protocol and generated-program resource for measuring when executable reasoning scaffolds improve SLM performance on MCQA tasks. CGR standardizes six components: a normalized item interface, a direct solver prompt, a generator prompt, a Python scaffold, solver-call and extraction helpers, and a three-channel result record. On 20,498 retained result rows from a locally prepared MCQA bundle and six metadata-registered solver models, the observed non-zero-baseline partition shows 66.21% macro assisted accuracy versus 38.11% direct accuracy, a +28.10 percentage-point difference with a pair-bootstrap interval of [20.32, 36.43]. Under a stricter Ab > 30% direct-signal gate, the macro difference is +14.11 points. These estimates are descriptive. Assisted inference uses a larger solver-call budget, answer extraction is brittle, Time-MQA contains the observed regressions, and some generated programs violate the no-hard-coding instruction. CGR provides the trace package needed to interpret these results, including direct, assisted, and generator-side answers, partition definitions, generated programs, response metadata, and audits.

Evaluating large language models (LLMs) in the biomedical domain requires benchmarks that can distinguish reasoning from pattern matching and remain discriminative as model capabilities improve. Existing biomedical question answering (QA) benchmarks are limited in this respect. Multiple-choice formats can allow models to succeed through answer elimination rather than inference, while widely circulated exam-style datasets are increasingly vulnerable to performance saturation and training data contamination. Multi-hop reasoning, defined as the ability to integrate information across multiple sources to derive an answer, is central to clinically meaningful tasks such as diagnostic support, literature-based discovery, and hypothesis generation, yet remains underrepresented in current biomedical QA benchmarks. MedHopQA is a disease-centered multi-hop reasoning benchmark consisting of 1,000 expert-curated question-answer pairs introduced as a shared task at BioCreative IX. Each question requires synthesis of information across two distinct Wikipedia articles, and answers are provided in an open-ended free-text format. Gold annotations are augmented with ontology-grounded synonym sets from MONDO, NCBI Gene, and NCBI Taxonomy to support both lexical and concept-level evaluation. MedHopQA was constructed through a structured process combining human annotation, triage, iterative verification, and LLM-as-a-judge validation. To reduce leaderboard gaming and contamination risk, the 1,000 scored questions are embedded within a publicly downloadable set of 10,000 questions, with answers withheld, on a CodaBench leaderboard. MedHopQA provides both a benchmark and a reusable framework for constructing future biomedical QA datasets that prioritize compositional reasoning, saturation resistance, and contamination resistance as core design constraints.

Long chain-of-thought (CoT) is an essential ingredient in effective usage of modern large language models, but our understanding of the reasoning strategies underlying these capabilities remains limited. While some prior works have attempted to categorize CoTs using predefined strategy types, such approaches are constrained by human intuition and fail to capture the full diversity of model behaviors. In this work, we introduce the CoT Encyclopedia, a bottom-up framework for analyzing and steering model reasoning. Our method automatically extracts diverse reasoning criteria from model-generated CoTs, embeds them into a semantic space, clusters them into representative categories, and derives contrastive rubrics to interpret reasoning behavior. Human evaluations show that this framework produces more interpretable and comprehensive analyses than existing methods. Moreover, we demonstrate that this understanding enables performance gains: we can predict which strategy a model is likely to use and guide it toward more effective alternatives. Finally, we provide practical insights, such as that training data format (e.g., free-form vs. multiple-choice) has a far greater impact on reasoning behavior than data domain, underscoring the importance of format-aware model design.

Large audio-language models (LALMs) have achieved near-human performance in sentence-level transcription and emotion recognition. However, existing evaluations focus mainly on surface-level perception, leaving the capacity of models for contextual and inference-driven reasoning in speech-based scenarios insufficiently examined. To address this gap, we introduce SpeechR, a unified benchmark for evaluating reasoning over speech in large audio-language models. SpeechR evaluates models along three key dimensions: factual retrieval, procedural inference, and normative judgment. It includes three distinct evaluation formats. The multiple-choice version measures answer selection accuracy. The generative version assesses the coherence and logical consistency of reasoning chains. The acoustic-feature version investigates whether variations in stress and emotion affect reasoning performance. Evaluations on eleven state-of-the-art LALMs reveal that high transcription accuracy does not translate into strong reasoning capabilities. SpeechR establishes a structured benchmark for evaluating reasoning in spoken language, enabling more targeted analysis of model capabilities across diverse dialogue-based tasks.

We present a new probing dataset named PROST: Physical Reasoning about Objects Through Space and Time. This dataset contains 18,736 multiple-choice questions made from 14 manually curated templates, covering 10 physical reasoning concepts. All questions are designed to probe both causal and masked language models in a zero-shot setting. We conduct an extensive analysis which demonstrates that state-of-the-art pretrained models are inadequate at physical reasoning: they are influenced by the order in which answer options are presented to them, they struggle when the superlative in a question is inverted (e.g., most <-> least), and increasing the amount of pretraining data and parameters only yields minimal improvements. These results provide support for the hypothesis that current pretrained models' ability to reason about physical interactions is inherently limited by a lack of real world experience. By highlighting these limitations, we hope to motivate the development of models with a human-like understanding of the physical world.

We introduce Social IQa, the first largescale benchmark for commonsense reasoning about social situations. Social IQa contains 38,000 multiple choice questions for probing emotional and social intelligence in a variety of everyday situations (e.g., Q: "Jordan wanted to tell Tracy a secret, so Jordan leaned towards Tracy. Why did Jordan do this?" A: "Make sure no one else could hear"). Through crowdsourcing, we collect commonsense questions along with correct and incorrect answers about social interactions, using a new framework that mitigates stylistic artifacts in incorrect answers by asking workers to provide the right answer to a different but related question. Empirical results show that our benchmark is challenging for existing question-answering models based on pretrained language models, compared to human performance (>20% gap). Notably, we further establish Social IQa as a resource for transfer learning of commonsense knowledge, achieving state-of-the-art performance on multiple commonsense reasoning tasks (Winograd Schemas, COPA).

Cognitive textual and visual reasoning tasks, such as puzzles, series, and analogies, demand the ability to quickly reason, decipher, and evaluate patterns both textually and spatially. While LLMs and VLMs, through extensive training on large amounts of human-curated data, have attained a high level of pseudo-human intelligence in some common sense reasoning tasks, they still struggle with more complex reasoning tasks that require cognitive understanding. In this work, we introduce a new dataset, NTSEBench, designed to evaluate the cognitive multi-modal reasoning and problem-solving skills of large models. The dataset comprises 2,728 multiple-choice questions comprising of a total of 4,642 images across 26 categories sampled from the NTSE examination conducted nationwide in India, featuring both visual and textual general aptitude questions that do not rely on rote learning. We establish baselines on the dataset using state-of-the-art LLMs and VLMs. To facilitate a comparison between open source and propriety models, we propose four distinct modeling strategies to handle different modalities (text and images) in the dataset instances.

Multimodal Large Language Models (MLLMs) have shown promising capabilities in mathematical reasoning within visual contexts across various datasets. However, most existing multimodal math benchmarks are limited to single-visual contexts, which diverges from the multi-visual scenarios commonly encountered in real-world mathematical applications. To address this gap, we introduce MV-MATH: a meticulously curated dataset of 2,009 high-quality mathematical problems. Each problem integrates multiple images interleaved with text, derived from authentic K-12 scenarios, and enriched with detailed annotations. MV-MATH includes multiple-choice, free-form, and multi-step questions, covering 11 subject areas across 3 difficulty levels, and serves as a comprehensive and rigorous benchmark for assessing MLLMs' mathematical reasoning in multi-visual contexts. Through extensive experimentation, we observe that MLLMs encounter substantial challenges in multi-visual math tasks, with a considerable performance gap relative to human capabilities on MV-MATH. Furthermore, we analyze the performance and error patterns of various models, providing insights into MLLMs' mathematical reasoning capabilities within multi-visual settings.

A benchmark score credits final answers, but not the route by which an item can be answered. In medical multimodal multiple-choice questions (MCQs), this distinction matters because a correct answer can be supported by the intended image finding or by benchmark-preserved cues in the wording of answers, non-visual clinical text, visible image text, artificial annotations, or device/context artifacts. We call the resulting score-level overinterpretation reasoning inflation. Here, a route is an observable input path that can support answer selection, not a claim about the model's hidden cognition. Across six medical multimodal MCQ datasets, we separate candidate cues from behavioral evidence through prompt- and image-side audits, modality ablations, and matched repairs that preserve the medical target and answer key. In a 13-configuration open-model panel, full-input accuracy is 62.63%, while text-only and options-only settings achieve 53.96% and 29.71%, respectively. Removing length-gap, absolute/conspicuous, and spatial/prepositional cues lowers accuracy by 6.58, 3.50, and 4.77 percentage points. We also construct MedQA-MM, a 1,000-item shortcut-mitigated subset, where text-only and options-only accuracy fall to 5.21% and 12.33%. This does not imply that models never use images; it shows that medical image-reasoning claims require route-level evidence.

Everyday conversations require understanding everyday events, which in turn, requires understanding temporal commonsense concepts interwoven with those events. Despite recent progress with massive pre-trained language models (LMs) such as T5 and GPT-3, their capability of temporal reasoning in dialogs remains largely under-explored. In this paper, we present the first study to investigate pre-trained LMs for their temporal reasoning capabilities in dialogs by introducing a new task and a crowd-sourced English challenge set, TIMEDIAL. We formulate TIME-DIAL as a multiple-choice cloze task with over 1.1K carefully curated dialogs. Empirical results demonstrate that even the best performing models struggle on this task compared to humans, with 23 absolute points of gap in accuracy. Furthermore, our analysis reveals that the models fail to reason about dialog context correctly; instead, they rely on shallow cues based on existing temporal patterns in context, motivating future research for modeling temporal concepts in text and robust contextual reasoning about them. The dataset is publicly available at: https://github.com/google-research-datasets/timedial.

Recent advancements in foundation models have enhanced AI systems' capabilities in autonomous tool usage and reasoning. However, their ability in location or map-based reasoning - which improves daily life by optimizing navigation, facilitating resource discovery, and streamlining logistics - has not been systematically studied. To bridge this gap, we introduce MapEval, a benchmark designed to assess diverse and complex map-based user queries with geo-spatial reasoning. MapEval features three task types (textual, API-based, and visual) that require collecting world information via map tools, processing heterogeneous geo-spatial contexts (e.g., named entities, travel distances, user reviews or ratings, images), and compositional reasoning, which all state-of-the-art foundation models find challenging. Comprising 700 unique multiple-choice questions about locations across 180 cities and 54 countries, MapEval evaluates foundation models' ability to handle spatial relationships, map infographics, travel planning, and navigation challenges. Using MapEval, we conducted a comprehensive evaluation of 28 prominent foundation models. While no single model excelled across all tasks, Claude-3.5-Sonnet, GPT-4o, and Gemini-1.5-Pro achieved competitive performance overall. However, substantial performance gaps emerged, particularly in MapEval, where agents with Claude-3.5-Sonnet outperformed GPT-4o and Gemini-1.5-Pro by 16% and 21%, respectively, and the gaps became even more amplified when compared to open-source LLMs. Our detailed analyses provide insights into the strengths and weaknesses of current models, though all models still fall short of human performance by more than 20% on average, struggling with complex map images and rigorous geo-spatial reasoning. This gap highlights MapEval's critical role in advancing general-purpose foundation models with stronger geo-spatial understanding.

Scientific research demands sophisticated reasoning over multimodal data, a challenge especially prevalent in biology. Despite recent advances in multimodal large language models (MLLMs) for AI-assisted research, existing multimodal reasoning benchmarks only target up to college-level difficulty, while research-level benchmarks emphasize lower-level perception, falling short of the complex multimodal reasoning needed for scientific discovery. To bridge this gap, we introduce MicroVQA, a visual-question answering (VQA) benchmark designed to assess three reasoning capabilities vital in research workflows: expert image understanding, hypothesis generation, and experiment proposal. MicroVQA consists of 1,042 multiple-choice questions (MCQs) curated by biology experts across diverse microscopy modalities, ensuring VQA samples represent real scientific practice. In constructing the benchmark, we find that standard MCQ generation methods induce language shortcuts, motivating a new two-stage pipeline: an optimized LLM prompt structures question-answer pairs into MCQs; then, an agent-based `RefineBot' updates them to remove shortcuts. Benchmarking on state-of-the-art MLLMs reveal a peak performance of 53\%; models with smaller LLMs only slightly underperform top models, suggesting that language-based reasoning is less challenging than multimodal reasoning; and tuning with scientific articles enhances performance. Expert analysis of chain-of-thought responses shows that perception errors are the most frequent, followed by knowledge errors and then overgeneralization errors. These insights highlight the challenges in multimodal scientific reasoning, showing MicroVQA is a valuable resource advancing AI-driven biomedical research. MicroVQA is available at https://huggingface.co/datasets/jmhb/microvqa, and project page at https://jmhb0.github.io/microvqa.

Long-form legal reasoning remains a key challenge for large language models (LLMs) in spite of recent advances in test-time scaling. We introduce LEXam, a novel benchmark derived from 340 law exams spanning 116 law school courses across a range of subjects and degree levels. The dataset comprises 4,886 law exam questions in English and German, including 2,841 long-form, open-ended questions and 2,045 multiple-choice questions. Besides reference answers, the open questions are also accompanied by explicit guidance outlining the expected legal reasoning approach such as issue spotting, rule recall, or rule application. Our evaluation on both open-ended and multiple-choice questions present significant challenges for current LLMs; in particular, they notably struggle with open questions that require structured, multi-step legal reasoning. Moreover, our results underscore the effectiveness of the dataset in differentiating between models with varying capabilities. Adopting an LLM-as-a-Judge paradigm with rigorous human expert validation, we demonstrate how model-generated reasoning steps can be evaluated consistently and accurately. Our evaluation setup provides a scalable method to assess legal reasoning quality beyond simple accuracy metrics. Project page: https://lexam-benchmark.github.io/

Spatial reasoning is a fundamental aspect of human cognition, yet it remains a major challenge for contemporary vision-language models (VLMs). Prior work largely relied on synthetic or LLM-generated environments with limited task designs and puzzle-like setups, failing to capture the real-world complexity, visual noise, and diverse spatial relationships that VLMs encounter. To address this, we introduce SpatiaLab, a comprehensive benchmark for evaluating VLMs' spatial reasoning in realistic, unconstrained contexts. SpatiaLab comprises 1,400 visual question-answer pairs across six major categories: Relative Positioning, Depth & Occlusion, Orientation, Size & Scale, Spatial Navigation, and 3D Geometry, each with five subcategories, yielding 30 distinct task types. Each subcategory contains at least 25 questions, and each main category includes at least 200 questions, supporting both multiple-choice and open-ended evaluation. Experiments across diverse state-of-the-art VLMs, including open- and closed-source models, reasoning-focused, and specialized spatial reasoning models, reveal a substantial gap in spatial reasoning capabilities compared with humans. In the multiple-choice setup, InternVL3.5-72B achieves 54.93% accuracy versus 87.57% for humans. In the open-ended setting, all models show a performance drop of around 10-25%, with GPT-5-mini scoring highest at 40.93% versus 64.93% for humans. These results highlight key limitations in handling complex spatial relationships, depth perception, navigation, and 3D geometry. By providing a diverse, real-world evaluation framework, SpatiaLab exposes critical challenges and opportunities for advancing VLMs' spatial reasoning, offering a benchmark to guide future research toward robust, human-aligned spatial understanding. SpatiaLab is available at: https://spatialab-reasoning.github.io/.

Natural language information needs over symbolic music scores rarely reduce to a single step lookup. Many queries require compositional Music Information Retrieval (MIR) that extracts multiple pieces of evidence from structured notation and aggregates them to answer the question. This setting remains challenging for Large Language Models due to the mismatch between natural language intents and symbolic representations, as well as the difficulty of reliably handling long structured contexts. Existing benchmarks only partially capture these retrieval demands, often emphasizing isolated theoretical knowledge or simplified settings. We introduce CSyMR-Bench, a benchmark for compositional MIR in symbolic music reasoning grounded in authentic user scenarios. It contains 126 multiple choice questions curated from community discussions and professional examinations, where each item requires chaining multiple atomic analyses over a score to derive implicit musical evidence. To support diagnosis, we provide a taxonomy with six query intent categories and six analytical dimension tags. We further propose a tool-augmented retrieval and reasoning framework that integrates a ReAct-style controller with deterministic symbolic analysis operators built with music21. Experiments across prompting baselines and agent variants show that tool-grounded compositional retrieval consistently outperforms Large Language Model-only approaches, yielding 5-7% absolute accuracy gains, with the largest improvements on analysis-heavy categories.

Recent advances have enabled unified omni-modal models in understanding audio, vision, and language. However, existing benchmarks, training data, and learning methods largely treat the modalities independently, leaving the capability of audio-visual joint reasoning poorly evaluated and insufficiently elicited. We address this gap with a benchmark, data engine, and learning method. First, we introduce OmniReasoningBench, a benchmark where both audio and visual evidence are indispensable. It comprises 1,150 multiple-choice and open-ended questions across two tasks, reasoning over video and reasoning beyond video. Second, we develop a data engine OmniQA. It automatically constructs evidence-grounded QA pairs that explicitly necessitate audio-visual joint reasoning, together with time-stamped clue chains that guide the annotation of thinking process. Besides our benchmark, this engine produces training data OmniReasoning-SFT-112K and OmniReasoning-RL-19K. Finally, we propose an on-policy self-distillation method Modality-Factored Self-Distillation (MFSD). It evaluates each sampled response under modality-specific clue contexts, disentangling the contributions of individual clues and their cross-modal interactions for token-level credit assignment. With our training data and learning method, our model OmniReasoning-30B-A3B achieves 50.0% on OmniVideoBench and 42.5% on OmniReasoningBench, improving the base model Qwen3-Omni-30B-A3B-Thinking by 12.8 and 9.3 percentage points, respectively. Moreover, it delivers substantial gains on general and long-video benchmarks, including Video-MME-v2. We hope our work offers a solid step for facilitating future research in omni-modal joint reasoning.

This paper introduces LongBench v2, a benchmark designed to assess the ability of LLMs to handle long-context problems requiring deep understanding and reasoning across real-world multitasks. LongBench v2 consists of 503 challenging multiple-choice questions, with contexts ranging from 8k to 2M words, across six major task categories: single-document QA, multi-document QA, long in-context learning, long-dialogue history understanding, code repository understanding, and long structured data understanding. To ensure the breadth and the practicality, we collect data from nearly 100 highly educated individuals with diverse professional backgrounds. We employ both automated and manual review processes to maintain high quality and difficulty, resulting in human experts achieving only 53.7% accuracy under a 15-minute time constraint. Our evaluation reveals that the best-performing model, when directly answers the questions, achieves only 50.1% accuracy. In contrast, the o1-preview model, which includes longer reasoning, achieves 57.7%, surpassing the human baseline by 4%. These results highlight the importance of enhanced reasoning ability and scaling inference-time compute to tackle the long-context challenges in LongBench v2. The project is available at https://longbench2.github.io.

Large vision-language models (LVLMs) augment language models with visual understanding, enabling multimodal reasoning. However, due to the modality gap between textual and visual data, they often face significant challenges, such as over-reliance on text priors, hallucinations, and limited capacity for complex visual reasoning. Existing benchmarks to evaluate visual reasoning in LVLMs often rely on schematic or synthetic images and on imprecise machine-generated explanations. To bridge the modality gap, we present DrivingVQA, a new benchmark derived from driving theory tests to evaluate visual chain-of-thought reasoning in complex real-world scenarios. It offers 3,931 expert-crafted multiple-choice problems and interleaved explanations grounded with entities relevant to the reasoning process. We leverage this dataset to perform an extensive study of LVLMs' ability to reason about complex visual scenarios. Our experiments reveal that open-source and proprietary LVLMs struggle with visual chain-of-thought reasoning under zero-shot settings. We investigate training strategies that leverage relevant entities to improve visual reasoning. Notably, we observe a performance boost of up to 7\% when reasoning over image tokens of cropped regions tied to these entities.

A fundamental aspect of compositional reasoning in a video is associating people and their actions across time. Recent years have seen great progress in general-purpose vision or video models and a move towards long-video understanding. While exciting, we take a step back and ask: are current models good at compositional reasoning on short videos? To this end, we introduce VELOCITI, a benchmark to study Video-LLMs by disentangling and assessing the comprehension of agents, actions, and their associations across multiple events. We adopt the Video-Language Entailment setup and propose StrictVLE that requires correct classification (rather than ranking) of the positive and negative caption. We evaluate several models and observe that even the best, LLaVA-OneVision (44.5%) and Gemini-1.5-Pro (49.3%), are far from human accuracy at 93.0%. Results show that action understanding lags behind agents, and negative captions created using entities appearing in the video perform worse than those obtained from pure text manipulation. We also present challenges with ClassicVLE and multiple-choice (MC) evaluation, strengthening our preference for StrictVLE. Finally, we validate that our benchmark requires visual inputs of multiple frames making it ideal to study video-language compositional reasoning.

We investigate how to teach large language models (LLMs) to perform scientific reasoning by leveraging expert discussions as a learning signal. Focusing on the genomics domain, we develop an automated pipeline to extract trainable data and introduce Genome-Bench, a new benchmark constructed from over a decade of scientific forum discussions on genome engineering. Our pipeline transforms raw interactions into a reinforcement learning-friendly multiple-choice questions format, supported by 3000+ high-quality question-answer pairs spanning foundational biology, experimental troubleshooting, tool usage, and beyond. We fine-tune an LLM using RL with a rule-based reward signal derived from the synthetic MCQ dataset to enhance domain-specific reasoning. Our results show that reinforcement learning from scientific discussions improves model performance by over 15% compared to the base model on Genome-Bench, narrowing the gap between open-source LLMs and expert-level reasoning. To our knowledge, this is the first end-to-end pipeline for teaching LLMs to reason from scientific discussions, with promising potential for generalization across scientific domains beyond biology.

Physics problem-solving is a challenging domain for large AI models, requiring integration of conceptual understanding, mathematical reasoning, and interpretation of physical diagrams. Current evaluation methodologies show notable limitations in capturing the breadth and complexity of undergraduate-level physics, underscoring the need for more rigorous assessments. To this end, we present PhysUniBench, a large-scale multimodal benchmark designed to evaluate and improve the reasoning capabilities of multimodal large language models (MLLMs) specifically on undergraduate-level physics problems. PhysUniBench consists of 3,304 physics questions spanning 8 major sub-disciplines of physics, each accompanied by one visual diagrams. The benchmark includes both open-ended and multiple-choice questions, systematically curated and difficulty-rated through an iterative model-in-the-loop process. The benchmark's construction involved a rigorous multi-stage process, including multiple roll-outs, expert-level evaluation, automated filtering of easily solved problems, and a nuanced difficulty grading system with five levels. Through extensive experiments, we observe that current state-of-the-art models encounter substantial challenges in physics reasoning. For example, GPT-4o mini achieves only about 34.2\% accuracy in the proposed PhysUniBench. These results highlight that current MLLMs struggle with advanced physics reasoning, especially on multi-step problems and those requiring precise diagram interpretation. By providing a broad and rigorous assessment tool, PhysUniBench aims to drive progress in AI for Science, encouraging the development of models with stronger physical reasoning, problem-solving skills, and multimodal understanding. The benchmark and evaluation scripts are available at https://prismax-team.github.io/PhysUniBenchmark/.

Large Multimodal Models (LMMs) have achieved remarkable progress across various capabilities; however, complex video reasoning in the scientific domain remains a significant and challenging frontier. Current video benchmarks predominantly target general scenarios where perception/recognition is heavily relied on, while with relatively simple reasoning tasks, leading to saturation and thus failing to effectively evaluate advanced multimodal cognitive skills. To address this critical gap, we introduce SciVideoBench, a rigorous benchmark specifically designed to assess advanced video reasoning in scientific contexts. SciVideoBench consists of 1,000 carefully crafted multiple-choice questions derived from cutting-edge scientific experimental videos spanning over 25 specialized academic subjects and verified by a semi-automatic system. Each question demands sophisticated domain-specific knowledge, precise spatiotemporal perception, and intricate logical reasoning, effectively challenging models' higher-order cognitive abilities. Our evaluation highlights significant performance deficits in state-of-the-art proprietary and open-source LMMs, including Gemini 2.5 Pro and Qwen2.5-VL, indicating substantial room for advancement in video reasoning capabilities. Detailed analyses of critical factors such as reasoning complexity and visual grounding provide valuable insights and clear direction for future developments in LMMs, driving the evolution of truly capable multimodal AI co-scientists. We hope SciVideoBench could fit the interests of the community and help to push the boundary of cutting-edge AI for border science.

While reasoning-enhanced large language models perform strongly on English medical tasks, a persistent multilingual gap remains, with substantially weaker reasoning in local languages, limiting equitable global medical deployment. To bridge this gap, we introduce Med-CoReasoner, a language-informed co-reasoning framework that elicits parallel English and local-language reasoning, abstracts them into structured concepts, and integrates local clinical knowledge into an English logical scaffold via concept-level alignment and retrieval. This design combines the structural robustness of English reasoning with the practice-grounded expertise encoded in local languages. To evaluate multilingual medical reasoning beyond multiple-choice settings, we construct MultiMed-X, a benchmark covering seven languages with expert-annotated long-form question answering and natural language inference tasks, comprising 350 instances per language. Experiments across three benchmarks show that Med-CoReasoner improves multilingual reasoning performance by an average of 5%, with particularly substantial gains in low-resource languages. Moreover, model distillation and expert evaluation analysis further confirm that Med-CoReasoner produces clinically sound and culturally grounded reasoning traces.

Despite state-of-the-art vision-language models (VLMs) have demonstrated strong reasoning capabilities, their performance in multilingual mathematical reasoning remains underexplored, particularly when compared to human performance. To bridge this gap, we introduce M3Kang, the first massively multilingual, multimodal mathematical reasoning dataset for VLMs. It is derived from the Kangaroo Math Competition, the world's largest mathematics contest, which annually engages over six million participants under the age of 18 across more than 90 countries. M3Kang includes 1,747 unique multiple-choice problems organized by grade-level difficulty, with translations into 108 culturally diverse languages, some of them including diagrams essential for solving them. Using this dataset, we conduct extensive benchmarking on both closed- and open-source SOTA models. We observe that, despite recent advances, models still struggle with basic math and diagram-based reasoning, with performance scaling with language presence and model size, but not with grade level. We also find that multilingual techniques can be effectively extended to the multimodal setting, resulting in significant improvements over baseline approaches. Our analysis also incorporates performance data from over 68,000 students, enabling direct comparison with human performance. We are open-sourcing M3Kang, including the English-only subset M2Kang, along with the framework and codebase used to construct the dataset.

Recent advancements in multimodal large language models (MLLMs) have shown strong understanding of driving scenes, drawing interest in their application to autonomous driving. However, high-level reasoning in safety-critical scenarios, where avoiding one traffic risk can create another, remains a major challenge. Such reasoning is often infeasible with only a single front view and requires a comprehensive view of the environment, which we achieve through multi-view inputs. We define Safety-Critical Reasoning as a new task that leverages multi-view inputs to address this challenge. Then, we distill Safety-Critical Reasoning into two stages: first resolve the immediate risk, then mitigate the decision-induced downstream risks. To support this, we introduce WaymoQA, a dataset of 35,000 human-annotated question-answer pairs covering complex, high-risk driving scenarios. The dataset includes multiple-choice and open-ended formats across both image and video modalities. Experiments reveal that existing MLLMs underperform in safety-critical scenarios compared to normal scenes, but fine-tuning with WaymoQA significantly improves their reasoning ability, highlighting the effectiveness of our dataset in developing safer and more reasoning-capable driving agents. Our code and data are provided in https://github.com/sjyu001/WaymoQA

Developing 3D vision-language models with robust clinical reasoning remains a challenge due to the inherent complexity of volumetric medical imaging, the tendency of models to overfit superficial report patterns, and the lack of interpretability-aware reward designs. In this paper, we propose Med3D-R1, a reinforcement learning framework with a two-stage training process: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). During SFT stage, we introduce a residual alignment mechanism to bridge the gap between high-dimensional 3D features and textual embeddings, and an abnormality re-weighting strategy to emphasize clinically informative tokens and reduce structural bias in reports. In RL stage, we redesign the consistency reward to explicitly promote coherent, step-by-step diagnostic reasoning. We evaluate our method on medical multiple-choice visual question answering using two 3D diagnostic benchmarks, CT-RATE and RAD-ChestCT, where our model attains state-of-the-art accuracies of 41.92\% on CT-RATE and 44.99\% on RAD-ChestCT. These results indicate improved abnormality diagnosis and clinical reasoning and outperform prior methods on both benchmarks. Overall, our approach holds promise for enhancing real-world diagnostic workflows by enabling more reliable and transparent 3D medical vision-language systems.

Real-world audio-visual understanding requires chaining evidence that is sparse, temporally dispersed, and split across the visual and auditory streams, whereas existing benchmarks largely fail to evaluate this capability. They restrict videos to short clips, isolate modalities, or reduce questions to one-hop perception. We introduce TraceAV-Bench, the first benchmark to jointly evaluate multi-hop reasoning over long audio-visual trajectories and multimodal hallucination robustness. TraceAV-Bench comprises 2,200 rigorously validated multiple-choice questions over 578 long videos, totaling 339.5 hours, spanning 4 evaluation dimensions and 15 sub-tasks. Each question is grounded in an explicit reasoning chain that averages 3.68 hops across a 15.1-minute temporal span. The dataset is built by a three-step semi-automated pipeline followed by a strict quality assurance process. Evaluation of multiple representative OmniLLMs on TraceAV-Bench reveals that the benchmark poses a persistent challenge across all models, with the strongest closed-source model (Gemini 3.1 Pro) reaching only 68.29% on general tasks, and the best open-source model (Ming-Flash-Omni-2.0) reaching 51.70%, leaving substantial headroom. Moreover, we find that robustness to multimodal hallucination is largely decoupled from general multimodal reasoning performance. We anticipate that TraceAV-Bench will stimulate further research toward OmniLLMs that can reason coherently and faithfully over long-form audio-visual content.

Longitudinal chest X-ray (CXR) interpretation requires reasoning over disease evolution across multiple patient visits, yet most existing medical VQA benchmarks focus on single images or short-horizon image pairs. We introduce MI-CXR, a benchmark for standardized evaluation of Multi-Interval longitudinal reasoning over multi-visit CXR sequences, without requiring free-form report generation or additional clinical context. MI-CXR comprises five-way multiple-choice questions over five-visit patient timelines and instantiates three complementary task families: Temporal Event Localization, Interval-wise Change Reasoning, and Global Trajectory Summarization, which assess clinically grounded visual reasoning over time. Evaluating 14 state-of-the-art vision-language models (VLMs) shows low overall performance, with an average accuracy of 29.3%, only modestly above random guessing. Using stage-wise diagnostic probing, we find that models often produce locally plausible interval descriptions but fail to enforce temporal constraints or compose evidence into globally consistent decisions over the full timeline. These findings reveal key limitations of current VLMs and establish MI-CXR as a principled benchmark for longitudinal medical reasoning. The benchmark is available at https://github.com/AIDASLab/MI-CXR

Recent advances in wearable sensing enable continuous monitoring of physiological and behavioral signals, yet existing benchmarks rarely evaluate whether AI systems can reason over a real user's longitudinal wearable record. We introduce WearableQA, a benchmark comprising 4,084 10-option multiple-choice questions constructed from the wearable time series, blood biomarkers, and demographics of 200 real users, each with up to 500 days of daily measurements. WearableQA preserves authentic wearable distributions that include device noise and inter-individual variability. To evaluate distinct reasoning capabilities, we introduce 16 question types organized along two complementary axes: data versus health reasoning, which distinguishes computation over longitudinal measurements from physiological interpretation; and single- versus cross-signal reasoning, which separates reasoning about individual signals from the integration of multiple signals. To construct reliable questions at scale, we adopt a dual-grounding framework that combines literature-grounded physiological findings with statistically validated population-grounded physiological patterns. This enables the capture of meaningful relationships observed in real-world wearable data. Evaluation of 14 proprietary and open-source LLMs demonstrates that WearableQA effectively differentiates model capabilities, with performance ranging from 19.6% to 72.9% against a 10% chance baseline. Moreover, WearableQA remains far from solved: most models achieve accuracies below 60%. Overall, WearableQA provides a realistic and diagnostic benchmark for evaluating LLM reasoning over real-world wearable data.

Chain-of-thought (CoT) instructions can distort multiple-choice VLM evaluation when a scorer appends a reasoning cue but reads answer-label logits before the model generates any rationale. We call this CoT-prefix scoring. On ScienceQA, Qwen2.5-VL-7B drops from 80.76% to 45.48%, and across five option-content permutations 93.54% of CoT-prefix predictions select the first slot. Condition-matched linear probes recover 78.94% from the same hidden states, while free generation restores 75.24%, showing that the answer often survives the prefix and the immediate readout fails. Vocabulary and layer diagnostics explain the mismatch: probability mass moves toward continuation tokens, while answer information remains linearly accessible in late layers. The effect recurs with varying severity across datasets and models, though not universally. These results show that CoT-prefix scoring can confound model knowledge with an evaluation-interface mismatch and should be avoided unless the requested and scored output events are aligned.

Medical reasoning models (MRMs) achieve superior performance on medical benchmarks compared to medical LLMs; however, high accuracy alone is insufficient for practical deployment. One of such requirements for real-world application is robustness to varying output constraints. Specifically, posing the same medical question while requesting different answer formats should not affect the underlying correctness of the response. We investigate this phenomenon in this paper, focusing on MRMs. To quantify this behavior, we propose the metric answer-format robustness: the ability to reliably generate correct outputs across varying specified formats. We examine three representative formats: multiple-choice, open-ended question-answering, and ranked lists. Across 15 proprietary and open-weight models, we observe substantial variation in format robustness (35-100%). Furthermore, we conduct controlled fine-tuning experiments on a shared backbone with matched training data to isolate the effects of the fine-tuning paradigm. We find that supervised fine-tuning yields more stable behavior across formats, whereas reinforcement fine-tuning often exhibits higher cross-format brittleness, with the degree of instability strongly dependent on reward design. Overall, answer-format robustness in MRMs is trainable yet brittle and requires careful evaluation for practical medical use.

Recent progress in large language model (LLM) reasoning has focused on domains like mathematics and coding, where abundant high-quality data and objective evaluation metrics are readily available. In contrast, progress in LLM reasoning models remains limited in scientific domains such as medicine and materials science due to limited dataset coverage and the inherent complexity of open-ended scientific questions. To address these challenges, we introduce WildSci, a new dataset of domain-specific science questions automatically synthesized from peer-reviewed literature, covering 9 scientific disciplines and 26 subdomains. By framing complex scientific reasoning tasks in a multiple-choice format, we enable scalable training with well-defined reward signals. We further apply reinforcement learning to finetune models on these data and analyze the resulting training dynamics, including domain-specific performance changes, response behaviors, and generalization trends. Experiments on a suite of scientific benchmarks demonstrate the effectiveness of our dataset and approach. We release WildSci to enable scalable and sustainable research in scientific reasoning, available at https://huggingface.co/datasets/JustinTX/WildSci.

As the usage of large language models for problems outside of simple text understanding or generation increases, assessing their abilities and limitations becomes crucial. While significant progress has been made in this area over the last few years, most research has focused on benchmarking English, leaving other languages underexplored. This makes evaluating the reasoning and robustness level of language models in Ukrainian particularly challenging. The purpose of this work is to establish a comprehensive benchmark for the reasoning capabilities evaluation of large language models in the Ukrainian language. This paper presents the ZNO-Eval benchmark based on real exam tasks from Ukraine's standardized educational testing system: the External Independent Evaluation and the National Multi-subject Test. With single-answer options, multiple-choice, matching, and open-ended questions from diverse subjects, including Ukrainian language, mathematics, history, and geography, this dataset paves the way toward a thorough analysis of reasoning capabilities across different domains and complexities. Evaluation of several well-known language models, such as GPT-3.5-Turbo, GPT-4o, GPT-4-Turbo, Mistral Large, Claude 3 Opus, and Gemini-1.5 Pro on this benchmark demonstrated the superiority of GPT-4o in both common knowledge reasoning and intricate language tasks. At the same time, Gemini Pro and GPT-4 Turbo excelled in the arithmetic domain, leading in single-answer and open-ended math problems. While all models were close to max performance in text-only common knowledge tasks like history and geography, there still is a gap for Ukrainian language and math, thus highlighting the importance of developing specialized language benchmarks for more accurate assessments of model capabilities and limitations across different languages and contexts.

Omni LLMs can naturally locate key audio and visual evidence and integrate information across modalities for audiovisual reasoning. Existing audio-visual reasoning benchmarks that mainly rely on multiple-choice questions usually score models solely based on answer correctness, making it difficult to assess whether a model truly forms a reasoning chain from evidence to the correct answer. In addition, textual or unimodal shortcuts may allow models to answer correctly without analyzing the necessary audio-visual evidence, further weakening the reliability of evaluation benchmarks. To address these issues, we introduce EvaOmni, an evidence-grounded benchmark containing 392 videos and 1,027 purely human-annotated question-answer pairs, covering 15 capabilities and three levels: retrieval and localization, basic understanding, and complex reasoning. We design a series of rigorous annotation procedures in which experts annotate audio and visual evidence and construct video-based distractors; humans review whether questions meet strict inspection criteria, and finally models filter textual and unimodal shortcuts. Beyond high-quality question-answer pairs, we require models to simultaneously return the answer and supporting evidence intervals in order to evaluate the complete reasoning chain. We propose the Coverage Grounded Score (C.G.) to measure how much annotated evidence a model identifies, and the Bottleneck Grounded Score (B.G.) to penalize poor temporal localization in either modality; both are incorporated into answer correctness. Across 7 open-weight and 17 proprietary Omni LLMs, Gemini3.7-Flash achieved the highest overall accuracy (79.4%), C.G. (70.8%), and B.G. (35.1%). At Level 3, it reached 79.1% accuracy, but its B.G. was only 12.3%. These results show that even the most advanced models still have room for development in grounded audio-visual reasoning.

We present Legal Argument Reasoning (LAR), a novel task designed to evaluate the legal reasoning capabilities of Large Language Models (LLMs). The task requires selecting the correct next statement (from multiple choice options) in a chain of legal arguments from court proceedings, given the facts of the case. We constructed a dataset (LAR-ECHR) for this task using cases from the European Court of Human Rights (ECHR). We evaluated seven general-purpose LLMs on LAR-ECHR and found that (a) the ranking of the models is aligned with that of LegalBench, an established US-based legal reasoning benchmark, even though LAR-ECHR is based on EU law, (b) LAR-ECHR distinguishes top models more clearly, compared to LegalBench, (c) even the best model (GPT-4o) obtains 75.8% accuracy on LAR-ECHR, indicating significant potential for further model improvement. The process followed to construct LAR-ECHR can be replicated with cases from other legal systems.

Reinforcement learning from verifiable rewards (RLVR) has recently gained attention for its ability to elicit self-evolved reasoning capabilitie from base language models without explicit reasoning supervisions, as demonstrated by DeepSeek-R1. While prior work on RLVR has primarily focused on mathematical and coding domains, its applicability to other tasks and domains remains unexplored. In this work, we investigate whether medical reasoning can emerge from RLVR. We introduce Med-RLVR as an initial study of RLVR in the medical domain leveraging medical multiple-choice question answering (MCQA) data as verifiable labels. Our results demonstrate that RLVR is not only effective for math and coding but also extends successfully to medical question answering. Notably, Med-RLVR achieves performance comparable to traditional supervised fine-tuning (SFT) on in-distribution tasks while significantly improving out-of-distribution generalization, with an 8-point accuracy gain. Further analysis of training dynamics reveals that, with no explicit reasoning supervision, reasoning emerges from the 3B-parameter base model. These findings underscore the potential of RLVR in domains beyond math and coding, opening new avenues for its application in knowledge-intensive fields such as medicine.

This work introduces MediQAl, a French medical question answering dataset designed to evaluate the capabilities of language models in factual medical recall and reasoning over real-world clinical scenarios. MediQAl contains 32,603 questions sourced from French medical examinations across 41 medical subjects. The dataset includes three tasks: (i) Multiple-Choice Question with Unique answer, (ii) Multiple-Choice Question with Multiple answer, and (iii) Open-Ended Question with Short-Answer. Each question is labeled as Understanding or Reasoning, enabling a detailed analysis of models' cognitive capabilities. We validate the MediQAl dataset through extensive evaluation with 14 large language models, including recent reasoning-augmented models, and observe a significant performance gap between factual recall and reasoning tasks. Our evaluation provides a comprehensive benchmark for assessing language models' performance on French medical question answering, addressing a crucial gap in multilingual resources for the medical domain.

The ability to reason from audio, including speech, paralinguistic cues, environmental sounds, and music, is essential for AI agents to interact effectively in real-world scenarios. Existing benchmarks mainly focus on static or single-scene settings and do not fully capture scenarios where multiple speakers, unfolding events, and heterogeneous audio sources interact. To address these challenges, we introduce MDAR, a benchmark for evaluating models on complex, multi-scene, and dynamically evolving audio reasoning tasks. MDAR comprises 3,000 carefully curated question-answer pairs linked to diverse audio clips, covering five categories of complex reasoning and spanning three question types. We benchmark 26 state-of-the-art audio language models on MDAR and observe that they exhibit limitations in complex reasoning tasks. On single-choice questions, Qwen2.5-Omni (open-source) achieves 76.67% accuracy, whereas GPT-4o Audio (closed-source) reaches 68.47%; however, GPT-4o Audio substantially outperforms Qwen2.5-Omni on the more challenging multiple-choice and open-ended tasks. Across all three question types, no model achieves 80% performance. These findings underscore the unique challenges posed by MDAR and its value as a benchmark for advancing audio reasoning research.Code and benchmark can be found at https://github.com/luckyerr/MDAR.

Large language models (LLMs) can achieve strong reasoning performance with sufficient computation, but they do not inherently know how much computation a task requires. We study budgeted inference-time reasoning for multiple tasks under a strict global token constraint and formalize it as a Ordered Stochastic Multiple-Choice Knapsack Problem(OS-MCKP). This perspective highlights a meta-cognitive requirement -- anticipating task difficulty, estimating return over investment (ROI), and allocating computation strategically. We propose ROI-Reasoning, a two-stage framework that endows LLMs with intrinsic, budget-aware rationality. In the first stage, Meta-Cognitive Fine-Tuning teaches models to predict reasoning cost and expected utility before generation, enabling explicit solve-or-skip decisions. Next, Rationality-Aware Reinforcement Learning optimizes sequential decision making under a hard token budget, allowing models to learn long-horizon allocation strategies. Across budgeted mathematical reasoning benchmarks, ROI-Reasoning consistently improves overall score while substantially reducing regret under tight computation budgets.

Recent progress in the reasoning capabilities of multimodal large language models (MLLMs) has empowered them to address more complex tasks such as scientific analysis and mathematical reasoning. Despite their promise, MLLMs' reasoning abilities across different scenarios in real life remain largely unexplored and lack standardized benchmarks for evaluation. To address this gap, we introduce MMR-Life, a comprehensive benchmark designed to evaluate the diverse multimodal multi-image reasoning capabilities of MLLMs across real-life scenarios. MMR-Life consists of 2,646 multiple-choice questions based on 19,108 images primarily sourced from real-world contexts, comprehensively covering seven reasoning types: abductive, analogical, causal, deductive, inductive, spatial, and temporal. Unlike existing reasoning benchmarks, MMR-Life does not rely on domain-specific expertise but instead requires models to integrate information across multiple images and apply diverse reasoning abilities. The evaluation of 37 advanced models highlights the substantial challenge posed by MMR-Life. Even top models like GPT-5 achieve only 58% accuracy and display considerable variance in performance across reasoning types. Moreover, we analyze the reasoning paradigms of existing MLLMs, exploring how factors such as thinking length, reasoning method, and reasoning type affect their performance. In summary, MMR-Life establishes a comprehensive foundation for evaluating, analyzing, and improving the next generation of multimodal reasoning systems.

Modeling plausible student misconceptions is critical for AI in education. In this work, we examine how large language models (LLMs) reason about misconceptions when generating multiple-choice distractors, a task that requires modeling incorrect yet plausible answers by coordinating solution knowledge, simulating student misconceptions, and evaluating plausibility. We introduce a taxonomy for analyzing the strategies used by state-of-the-art LLMs, examining their reasoning procedures and comparing them to established best practices in the learning sciences. Our structured analysis reveals a surprising alignment between their processes and best practices: the models typically solve the problem correctly first, then articulate and simulate multiple potential misconceptions, and finally select a set of distractors. An analysis of failure modes reveals that errors arise primarily from failures in recovering the correct solution and selecting among response candidates, rather than simulating errors or structuring the process. Consistent with these results, we find that providing the correct solution in the prompt improves alignment with human-authored distractors by 8%, highlighting the critical role of anchoring to the correct solution when generating plausible incorrect student reasoning. Overall, our analysis offers a structured and interpretable lens into LLMs' ability to model incorrect student reasoning and produce high-quality distractors.

Reasoning abilities of LLMs have been a key focus in recent years. One challenging reasoning domain with interesting nuances is legal reasoning, which requires careful application of rules, and precedents while balancing deductive and analogical reasoning, and conflicts between rules. Although there have been a few works on using LLMs for legal reasoning, their focus has been on overall accuracy. In this paper, we dig deeper to do a step-by-step analysis and figure out where they commit errors. We use the college-level Multiple Choice Question-Answering (MCQA) task from the Civil Procedure dataset and propose a new error taxonomy derived from initial manual analysis of reasoning chains with respect to several LLMs, including two objective measures: soundness and correctness scores. We then develop an LLM-based automated evaluation framework to identify reasoning errors and evaluate the performance of LLMs. The computation of soundness and correctness on the dataset using the auto-evaluator framework reveals several interesting insights. Furthermore, we show that incorporating the error taxonomy as feedback in popular prompting techniques marginally increases LLM performance. Our work will also serve as an evaluation framework that can be used in detailed error analysis of reasoning chains for logic-intensive complex tasks.

Large language models (LLMs) are entering legal workflows, yet we lack a jurisdiction-specific framework to assess their baseline competence therein. We use India's public legal examinations as a transparent proxy. Our multi-year benchmark assembles objective screens from top national and state exams and evaluates open and frontier LLMs under real-world exam conditions. To probe beyond multiple-choice questions, we also include a lawyer-graded, paired-blinded study of long-form answers from the Supreme Court's Advocate-on-Record exam. This is, to our knowledge, the first exam-grounded, India-specific yardstick for LLM court-readiness released with datasets and protocols. Our work shows that while frontier systems consistently clear historical cutoffs and often match or exceed recent top-scorer bands on objective exams, none surpasses the human topper on long-form reasoning. Grader notes converge on three reliability failure modes: procedural or format compliance, authority or citation discipline, and forum-appropriate voice and structure. These findings delineate where LLMs can assist (checks, cross-statute consistency, statute and precedent lookups) and where human leadership remains essential: forum-specific drafting and filing, procedural and relief strategy, reconciling authorities and exceptions, and ethical, accountable judgment.

Despite the remarkable coherence of Large Language Models (LLMs), existing evaluation methods often suffer from fluency bias and rely heavily on multiple-choice formats, making it difficult to assess factual accuracy and complex reasoning effectively. LLMs thus frequently generate factually inaccurate responses, especially in complex reasoning tasks, highlighting two prominent challenges: (1) the inadequacy of existing methods to evaluate reasoning and factual accuracy effectively, and (2) the reliance on human evaluators for nuanced judgment, as illustrated by Williams and Huckle (2024)[1], who found manual grading indispensable despite automated grading advancements. To address evaluation gaps in open-ended reasoning tasks, we introduce the EQUATOR Evaluator (Evaluation of Question Answering Thoroughness in Open-ended Reasoning). This framework combines deterministic scoring with a focus on factual accuracy and robust reasoning assessment. Using a vector database, EQUATOR pairs open-ended questions with human-evaluated answers, enabling more precise and scalable evaluations. In practice, EQUATOR significantly reduces reliance on human evaluators for scoring and improves scalability compared to Williams and Huckle's (2004)[1] methods. Our results demonstrate that this framework significantly outperforms traditional multiple-choice evaluations while maintaining high accuracy standards. Additionally, we introduce an automated evaluation process leveraging smaller, locally hosted LLMs. We used LLaMA 3.2B, running on the Ollama binaries to streamline our assessments. This work establishes a new paradigm for evaluating LLM performance, emphasizing factual accuracy and reasoning ability, and provides a robust methodological foundation for future research.

Enhancing reasoning capabilities in small language models is critical for specialized applications such as medical question answering, particularly in underrepresented languages like Persian. In this study, we employ Reinforcement Learning with AI Feedback (RLAIF) and Direct preference optimization (DPO) to improve the reasoning skills of a general-purpose Persian language model. To achieve this, we translated a multiple-choice medical question-answering dataset into Persian and used RLAIF to generate rejected-preferred answer pairs, which are essential for DPO training. By prompting both teacher and student models to produce Chain-of-Thought (CoT) reasoning responses, we compiled a dataset containing correct and incorrect reasoning trajectories. This dataset, comprising 2 million tokens in preferred answers and 2.5 million tokens in rejected ones, was used to train a baseline model, significantly enhancing its medical reasoning capabilities in Persian. Remarkably, the resulting model outperformed its predecessor, gaokerena-V, which was trained on approximately 57 million tokens, despite leveraging a much smaller dataset. These results highlight the efficiency and effectiveness of reasoning-focused training approaches in developing domain-specific language models with limited data availability.

It has been shown that Large Reasoning Models (LRMs) may not *say what they think*: they do not always volunteer information about how certain parts of the input influence their reasoning. But it is one thing for a model to *omit* such information and another, worse thing to *lie* about it. Here, we extend the work of Chen et al. (2025) to show that LRMs will do just this: they will flatly deny relying on hints provided in the prompt in answering multiple choice questions -- even when directly asked to reflect on unusual (i.e. hinted) prompt content, even when allowed to use hints, and even though experiments *show* them to be using the hints. Our results thus have discouraging implications for CoT monitoring and interpretability.

Reasoning-focused Question Answering (QA) has advanced rapidly with Large Language Models (LLMs), yet high-quality benchmarks for low-resource languages remain scarce. Persian, spoken by roughly 130 million people, lacks a comprehensive open-domain resource for evaluating reasoning-capable QA systems. We introduce PARSE, the first open-domain Persian reasoning QA benchmark, containing 10,800 questions across Boolean, multiple-choice, and factoid formats, with diverse reasoning types, difficulty levels, and answer structures. The benchmark is built via a controlled LLM-based generation pipeline and validated through human evaluation. We also ensure linguistic and factual quality through multi-stage filtering, annotation, and consistency checks. We benchmark multilingual and Persian LLMs under multiple prompting strategies and show that Persian prompts and structured prompting (CoT for Boolean/multiple-choice; few-shot for factoid) improve performance. Fine-tuning further boosts results, especially for Persian-specialized models. These findings highlight how PARSE supports both fair comparison and practical model adaptation. PARSE fills a critical gap in Persian QA research and provides a strong foundation for developing and evaluating reasoning-capable LLMs in low-resource settings.

Islamic inheritance law ('ilm al-mawarith) is challenging for large language models because solving inheritance cases requires complex, structured multi-step reasoning and the correct application of juristic rules to compute heirs' shares. We introduce MAWARITH, a large-scale annotated dataset of 12,500 Arabic inheritance cases for training and evaluating models on the full reasoning chain: (i) identifying eligible heirs, (ii) applying blocking (hajb) and allocation rules, and (iii) computing exact inheritance shares. Unlike prior datasets that restrict inheritance case solving to multiple-choice questions, MAWARITH supports the full reasoning chain and provides step-by-step solutions, including intermediate legal decisions and justifications based on classical juristic sources and established inheritance rules, as well as exact share calculations. To evaluate models beyond final-answer accuracy, we propose MIR-E (Mawarith Inheritance Reasoning Evaluation), a weighted multi-stage metric that scores key reasoning stages and captures error propagation across the pipeline. We evaluate six LLMs in a zero-shot setting. Gemini-2.5-flash achieves about 90% MIR-E on both validation and test, while Fanar-C, Fanar-Sadiq, LLaMA 3, and Qwen 3 remain below 50%. Our error analysis identifies recurring failure patterns, including scenario misinterpretation, errors in heir identification, errors in share allocation, and missing or incorrect application of key inheritance rules such as 'awl and radd. The MAWARITH dataset is publicly available at https://gitlab.com/islamgpt1/qias_shared_task_2026.

Multi-step symbolic reasoning is essential for robust financial analysis, yet most benchmarks neglect intermediate reasoning steps. FINCHAIN introduced verifiable Chain-of-Thought (CoT) evaluation but is limited to English. FINESSE-Bench includes a Russian block but relies on multiple-choice questions without step-level supervision. We present RusFinChain, the first Russian-language symbolic benchmark for verifiable CoT reasoning in finance. It spans 17 domains, 172 topics, and comprises 5,280 parameterized examples from executable Python templates, ensuring contamination-free evaluation. Each example includes a gold-standard reasoning chain with intermediate numeric values for automatic verification. We also introduce enhanced metrics: Fuzzy Numeric Alignment and Soft-Attention Alignment. We evaluate 8 open-weight LLMs on a stratified sample, generating 8,100 responses. Results reveal a substantial reasoning gap: models achieve Hard F1 of ~0.65 for step alignment, but only ~29% of final answers are correct. Our fuzzy and soft metrics show stronger correlation with final-answer correctness (Spearman rho approx 0.48) than the original ChainEval (rho approx 0.38-0.46), demonstrating superior diagnostic power. We release dataset, code, and evaluation framework to foster verifiable financial AI for the Russian-speaking community.

Recent advances in large language models (LLMs) have improved reasoning in text and image domains, yet achieving robust video reasoning remains a significant challenge. Existing video benchmarks mainly assess shallow understanding and reasoning and allow models to exploit global context, failing to rigorously evaluate true causal and stepwise reasoning. We present CausalStep, a benchmark designed for explicit stepwise causal reasoning in videos. CausalStep segments videos into causally linked units and enforces a strict stepwise question-answer (QA) protocol, requiring sequential answers and preventing shortcut solutions. Each question includes carefully constructed distractors based on error type taxonomy to ensure diagnostic value. The benchmark features 100 videos across six categories and 1,852 multiple-choice QA pairs. We introduce seven diagnostic metrics for comprehensive evaluation, enabling precise diagnosis of causal reasoning capabilities. Experiments with leading proprietary and open-source models, as well as human baselines, reveal a significant gap between current models and human-level stepwise reasoning. CausalStep provides a rigorous benchmark to drive progress in robust and interpretable video reasoning.

Large language models (LLMs) increasingly underpin scientific AI applications that reason over structured knowledge, from biomedical question answering to materials informatics. However, their logical reasoning often falls short, producing factual inaccuracies unacceptable in these settings. Reliable evaluation remains challenging: manual dataset construction scales poorly, and LLM-based generation risks embedding the very flaws it aims to measure. High-quality benchmarks must ground both correct and incorrect labelled examples in explicit background knowledge, formally verifiable by a standard reasoner. We propose a pipeline that automatically generates ontology-grounded multiple-choice question (MCQ) benchmarks from any sufficiently axiomatised OWL 2 ontology, with correct answers grounded in the ontology by design. Distractors are generated by perturbing the right-hand-side class expressions of class definition axioms, and their incorrectness is formally verified by an OWL reasoner via entailment checks. We evaluate the pipeline on three ontologies: Pizza (small, academic), PMDco (complex, materials science), and DOID (large, biomedical), generating 112, 2,491, and 15,216 MCQs respectively. Distractors span four semantic categories from class unsatisfiability to weakened subsumptions, enabling diagnostic evaluation of specific reasoning failures. Items meet natural language quality standards: mean LLM judge scores of 4.02, 4.36, and 3.36 out of 5 confirm fluency, and correct-answer-to-distractor similarity above 0.8 shows that wrong options cannot be dismissed on surface form alone. Six LLMs evaluated zero-shot achieve 41.1-76.8% accuracy, well above the 25% random-guessing baseline, indicating the benchmarks are challenging and discriminative. This work is a step towards more reliable benchmarks for assessing logical reasoning in scientific AI.

Abductive reasoning is inference to the most plausible explanation. For example, if Jenny finds her house in a mess when she returns from work, and remembers that she left a window open, she can hypothesize that a thief broke into her house and caused the mess, as the most plausible explanation. While abduction has long been considered to be at the core of how people interpret and read between the lines in natural language (Hobbs et al., 1988), there has been relatively little research in support of abductive natural language inference and generation. We present the first study that investigates the viability of language-based abductive reasoning. We introduce a challenge dataset, ART, that consists of over 20k commonsense narrative contexts and 200k explanations. Based on this dataset, we conceptualize two new tasks -- (i) Abductive NLI: a multiple-choice question answering task for choosing the more likely explanation, and (ii) Abductive NLG: a conditional generation task for explaining given observations in natural language. On Abductive NLI, the best model achieves 68.9% accuracy, well below human performance of 91.4%. On Abductive NLG, the current best language generators struggle even more, as they lack reasoning capabilities that are trivial for humans. Our analysis leads to new insights into the types of reasoning that deep pre-trained language models fail to perform--despite their strong performance on the related but more narrowly defined task of entailment NLI--pointing to interesting avenues for future research.

Large-scale sequence-to-sequence models have shown to be adept at both multiple-choice and open-domain commonsense reasoning tasks. However, the current systems do not provide the ability to control the various attributes of the reasoning chain. To enable better controllability, we propose to study the commonsense reasoning as a template filling task (TemplateCSR) -- where the language models fills reasoning templates with the given constraints as control factors. As an approach to TemplateCSR, we (i) propose a dataset of commonsense reasoning template-expansion pairs and (ii) introduce POTTER, a pretrained sequence-to-sequence model using prompts to perform commonsense reasoning across concepts. Our experiments show that our approach outperforms baselines both in generation metrics and factuality metrics. We also present a detailed error analysis on our approach's ability to reliably perform commonsense reasoning.

Multi-modal large language models(MLLMs) have achieved remarkable progress and demonstrated powerful knowledge comprehension and reasoning abilities. However, the mastery of domain-specific knowledge, which is essential for evaluating the intelligence of MLLMs, continues to be a challenge. Current multi-modal benchmarks for domain-specific knowledge concentrate on multiple-choice questions and are predominantly available in English, which imposes limitations on the comprehensiveness of the evaluation. To this end, we introduce CMMU, a novel benchmark for multi-modal and multi-type question understanding and reasoning in Chinese. CMMU consists of 3,603 questions in 7 subjects, covering knowledge from primary to high school. The questions can be categorized into 3 types: multiple-choice, multiple-response, and fill-in-the-blank, bringing greater challenges to MLLMs. In addition, we propose a rigorous evaluation strategy called ShiftCheck for assessing multiple-choice questions. The strategy aims to reduce position bias, minimize the influence of randomness on correctness, and perform a quantitative analysis of position bias. We evaluate seven open-source MLLMs along with GPT4-V, Gemini-Pro, and Qwen-VL-Plus. The results demonstrate that CMMU poses a significant challenge to the recent MLLMs.

Question Answering (QA) datasets have been instrumental in developing and evaluating Large Language Model (LLM) capabilities. However, such datasets are scarce for languages other than English due to the cost and difficulties of collection and manual annotation. This means that producing novel models and measuring the performance of multilingual LLMs in low-resource languages is challenging. To mitigate this, we propose SynDARin, a method for generating and validating QA datasets for low-resource languages. We utilize parallel content mining to obtain human-curated paragraphs between English and the target language. We use the English data as context to generate synthetic multiple-choice (MC) question-answer pairs, which are automatically translated and further validated for quality. Combining these with their designated non-English human-curated paragraphs form the final QA dataset. The method allows to maintain the content quality, reduces the likelihood of factual errors, and circumvents the need for costly annotation. To test the method, we created a QA dataset with 1.2K samples for the Armenian language. The human evaluation shows that 98% of the generated English data maintains quality and diversity in the question types and topics, while the translation validation pipeline can filter out sim70% of data with poor quality. We use the dataset to benchmark state-of-the-art LLMs, showing their inability to achieve human accuracy with some model performances closer to random chance. This shows that the generated dataset is non-trivial and can be used to evaluate reasoning capabilities in low-resource language.

Large language models (LLMs) have demonstrated solid zero-shot reasoning capabilities, which is reflected in their performance on the current test tasks. This calls for a more challenging benchmark requiring highly advanced reasoning ability to be solved. In this paper, we introduce such a benchmark, consisting of 191 long-form (1200 words on average) mystery narratives constructed as detective puzzles. Puzzles are sourced from the "5 Minute Mystery" platform and include a multiple-choice question for evaluation. Only 47% of humans solve a puzzle successfully on average, while the best human solvers achieve over 80% success rate. We show that GPT-3 models barely outperform random on this benchmark (with 28% accuracy) while state-of-the-art GPT-4 solves only 38% of puzzles. This indicates that there is still a significant gap in the deep reasoning abilities of LLMs and humans and highlights the need for further research in this area. Our work introduces a challenging benchmark for future studies on reasoning in language models and contributes to a better understanding of the limits of LLMs' abilities.

Multimodal Large Language Models are increasingly applied to biomedical imaging, yet scientific reasoning for microscopy remains limited by the scarcity of large-scale, high-quality training data. We introduce MicroVQA++, a three-stage, large-scale and high-quality microscopy VQA corpus derived from the BIOMEDICA archive. Stage one bootstraps supervision from expert-validated figure-caption pairs sourced from peer-reviewed articles. Stage two applies HiCQA-Graph, a novel heterogeneous graph over images, captions, and QAs that fuses NLI-based textual entailment, CLIP-based vision-language alignment, and agent signals to identify and filter inconsistent samples. Stage three uses a MultiModal Large Language Model (MLLM) agent to generate multiple-choice questions (MCQ) followed by human screening. The resulting release comprises a large training split and a human-checked test split whose Bloom's level hard-sample distribution exceeds the MicroVQA benchmark. Our work delivers (i) a quality-controlled dataset that couples expert literature with graph-based filtering and human refinement; (ii) HiCQA-Graph, the first graph that jointly models (image, caption, QA) for cross-modal consistency filtering; (iii) evidence that careful data construction enables 4B-scale MLLMs to reach competitive microscopy reasoning performance (e.g., GPT-5) and achieve state-of-the-art performance among open-source MLLMs. Code and dataset will be released after the review process concludes.

Recent Multimodal Large Language Models (MLLMs) achieve promising performance on visual and audio benchmarks independently. However, the ability of these models to process cross-modal information synchronously remains largely unexplored. We introduce Daily-Omni, a multiple-choice Audio-Visual QA benchmark featuring 684 real-world videos and 1,197 questions spanning 6 task families that explicitly require cross-modal temporal reasoning. To support scalable benchmark construction, we develop a semi-automatic pipeline for annotation, cross-modal consistency refinement, temporal alignment elicitation, and text-only leakage filtering, followed by human verification. We further provide a diagnostic evaluation suite and extensively evaluate 24 foundation models under 37 model--modality settings (Audio+Video / Audio-only / Video-only / Text-only). Finally, we include a training-free modular diagnostic baseline that composes off-the-shelf unimodal models to serve as a diagnostic baseline and to illustrate how explicit temporal alignment signals affect performance. Results indicate that many end-to-end MLLMs still struggle on alignment-critical questions, suggesting that robust cross-modal temporal alignment remains an important open challenge.

Islamic inheritance law (Ilm al-Mawarith) requires precise identification of heirs and calculation of shares, which poses a challenge for AI. In this paper, we present a lightweight framework for solving multiple-choice inheritance questions using a specialised Arabic text encoder and Attentive Relevance Scoring (ARS). The system ranks answer options according to semantic relevance, and enables fast, on-device inference without generative reasoning. We evaluate Arabic encoders (MARBERT, ArabicBERT, AraBERT) and compare them with API-based LLMs (Gemini, DeepSeek) on the QIAS 2025 dataset. While large models achieve an accuracy of up to 87.6%, they require more resources and are context-dependent. Our MARBERT-based approach achieves 69.87% accuracy, presenting a compelling case for efficiency, on-device deployability, and privacy. While this is lower than the 87.6% achieved by the best-performing LLM, our work quantifies a critical trade-off between the peak performance of large models and the practical advantages of smaller, specialized systems in high-stakes domains.

Recent advances in Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities across various vision-language tasks. However, their reasoning abilities in the multimodal symbolic music domain remain largely unexplored. We introduce WildScore, the first in-the-wild multimodal symbolic music reasoning and analysis benchmark, designed to evaluate MLLMs' capacity to interpret real-world music scores and answer complex musicological queries. Each instance in WildScore is sourced from genuine musical compositions and accompanied by authentic user-generated questions and discussions, capturing the intricacies of practical music analysis. To facilitate systematic evaluation, we propose a systematic taxonomy, comprising both high-level and fine-grained musicological ontologies. Furthermore, we frame complex music reasoning as multiple-choice question answering, enabling controlled and scalable assessment of MLLMs' symbolic music understanding. Empirical benchmarking of state-of-the-art MLLMs on WildScore reveals intriguing patterns in their visual-symbolic reasoning, uncovering both promising directions and persistent challenges for MLLMs in symbolic music reasoning and analysis. We release the dataset and code.

We propose Perceptual Taxonomy, a structured process of scene understanding that first recognizes objects and their spatial configurations, then infers task-relevant properties such as material, affordance, function, and physical attributes to support goal-directed reasoning. While this form of reasoning is fundamental to human cognition, current vision-language benchmarks lack comprehensive evaluation of this ability and instead focus on surface-level recognition or image-text alignment. To address this gap, we introduce Perceptual Taxonomy, a benchmark for physically grounded visual reasoning. We annotate 3173 objects with four property families covering 84 fine-grained attributes. Using these annotations, we construct a multiple-choice question benchmark with 5802 images across both synthetic and real domains. The benchmark contains 28033 template-based questions spanning four types (object description, spatial reasoning, property matching, and taxonomy reasoning), along with 50 expert-crafted questions designed to evaluate models across the full spectrum of perceptual taxonomy reasoning. Experimental results show that leading vision-language models perform well on recognition tasks but degrade by 10 to 20 percent on property-driven questions, especially those requiring multi-step reasoning over structured attributes. These findings highlight a persistent gap in structured visual understanding and the limitations of current models that rely heavily on pattern matching. We also show that providing in-context reasoning examples from simulated scenes improves performance on real-world and expert-curated questions, demonstrating the effectiveness of perceptual-taxonomy-guided prompting.

Large Vision-Language Models (LVLMs) excel in multimodal reasoning and have shown impressive performance on various multimodal benchmarks. However, most of these benchmarks evaluate models primarily through multiple-choice or short-answer formats, which do not take the reasoning process into account. Although some benchmarks assess the reasoning process, their methods are often overly simplistic and only examine reasoning when answers are incorrect. This approach overlooks scenarios where flawed reasoning leads to correct answers. In addition, these benchmarks do not consider the impact of intermodal relationships on reasoning. To address this issue, we propose the Reasoning Process Tree Score (RPTS), a tree structure-based metric to assess reasoning processes. Specifically, we organize the reasoning steps into a reasoning tree and leverage its hierarchical information to assign weighted faithfulness scores to each reasoning step. By dynamically adjusting these weights, RPTS not only evaluates the overall correctness of the reasoning, but also pinpoints where the model fails in the reasoning. To validate RPTS in real-world multimodal scenarios, we construct a new benchmark, RPTS-Eval, comprising 374 images and 390 reasoning instances. Each instance includes reliable visual-textual clues that serve as leaf nodes of the reasoning tree. Furthermore, we define three types of intermodal relationships to investigate how intermodal interactions influence the reasoning process. We evaluated representative LVLMs (e.g., GPT4o, Llava-Next), uncovering their limitations in multimodal reasoning and highlighting the differences between open-source and closed-source commercial LVLMs. We believe that this benchmark will contribute to the advancement of research in the field of multimodal reasoning.

Training expert LLMs in domains with scarce data is difficult, often relying on multiple-choice questions (MCQs). However, standard outcome-based reinforcement learning (RL) on MCQs is risky. While it may improve accuracy, we observe it often degrades reasoning quality such as logical consistency. Existing solutions to supervise reasoning, such as large-scale Process Reward Models (PRMs), are prohibitively expensive. To address this, we propose CLARity, a cost-effective RL framework that enhances reasoning quality using only a small, general-purpose LLM. CLARity integrates a consistency-aware reward mechanism with a 2-stage refine-then-monitor training pipeline to enhance reasoning consistency, and a dynamic data reformulation strategy to to better exploit limited data. Experiments demonstrate that CLARity improves response consistency by 16.5% and accuracy by 7.5% over baselines. Human evaluations further confirm holistic improvements in coherence and professionalism. Thus, CLARity offers a generalizable solution that enables smaller models to effectively guide expert models by reasoning consistency.Our code is open sourced at: https://github.com/Infinite-set/CLARity

Large Vision Language Models (LVLMs) possess extensive text knowledge but struggles to utilize this knowledge for fine-grained image recognition, often failing to differentiate between visually similar categories. Existing fine-tuning methods using Reinforcement Learning (RL) with exact-match reward signals are often brittle, encourage memorization of training categories, and fail to elicit differential reasoning needed for generalization to unseen classes. To address this, we propose DiVE-k, Differential Visual rEasoning using top-k generations, framework that leverages model's own top-k predictions as a training signal. For each training image, DiVE-k creates a multiple-choice question from the model's top-k outputs and uses RL to train the model to select the correct answer. This approach requires the model to perform fine-grained differential reasoning among plausible options and provides a simple, verifiable reward signal that mitigates memorization and improves generalization. Experiments on five standard fine-grained datasets show that our method significantly outperforms existing approaches. In the standard base-to-novel generalization setting, DiVE-k surpasses the QWEN2.5-VL-7B and ViRFT by 10.04% and 6.16% on the Harmonic Mean metric, respectively. Further experiments show similar gains in mixed-domain and few-shot scenarios.

Reasoning-focused large language models (LLMs) have advanced in many NLP tasks, yet their evaluation remains challenging: final answers alone do not expose the intermediate reasoning steps, making it difficult to determine whether a model truly reasons correctly and where failures occur, while existing multi-hop QA benchmarks lack step-level annotations for diagnosing reasoning failures. To address this gap, we propose Omanic, an open-domain multi-hop QA resource that provides decomposed sub-questions and intermediate answers as structural annotations for analyzing reasoning processes. It contains 10,296 machine-generated training examples (OmanicSynth) and 967 expert-reviewed human-annotated evaluation examples (OmanicBench). Systematic evaluations show that state-of-the-art LLMs achieve only 73.11% multiple-choice accuracy on OmanicBench, confirming its high difficulty. Stepwise analysis reveals that CoT's performance hinges on factual completeness, with its gains diminishing under knowledge gaps and errors amplifying in later hops. Additionally, supervised fine-tuning on OmanicSynth brings substantial transfer gains (7.41 average points) across six reasoning and math benchmarks, validating the dataset's quality and further supporting the effectiveness of OmanicSynth as supervision for reasoning-capability transfer. We release the data at https://huggingface.co/datasets/li-lab/Omanic and the code at https://github.com/XiaojieGu/Omanic.

Do Large Language Models (LLMs) possess genuine structural reasoning, or merely rely on surface-level pattern matching? The financial domain, demanding numerical precision and multi-step logic over long contexts, is an ideal testbed. Existing benchmarks fail to capture real-world industrial complexity, predominantly relying on multiple-choice questions or single-hop QA over cropped tables while ignoring intricate cross-statement dynamics and temporal de-cumulation. To bridge this gap, we introduce FinIndices, a large-scale benchmark evaluating data-processing fidelity over uncropped financial statements (up to 32K tokens). Utilizing an automated synthesis pipeline with adversarial traps, FinIndices encompasses Single-Index computation and Table-Index tabulation to test complex domain, temporal, and caliber reasoning. Our evaluation reveals two severe LLM vulnerabilities. First, a "Knowledge Bottleneck": despite memorizing formulas during pre-training, models demonstrate fragile pattern matching. Removing explicit formula hints causes performance to collapse (e.g., Gemini-3.1-Pro drops from 70.70% to 38.22% on table tasks), exposing fatal flaws in temporal de-cumulation and stock-flow caliber mismatch. Second, a "Structural Bottleneck": the intense cognitive load of generating multi-metric, multi-period tables actively drains reasoning capacity. Under structural pressure, LLMs that flawlessly execute isolated derivations regress to shallow heuristics, such as fetching incorrect adjacent columns or substituting deep accounting adjustments with lazy literal arithmetic. Finally, Supervised Fine-Tuning (SFT) yields substantial zero-hint gains (+8.54% Single, +3.82% Table), validating that structured logic can be partially restored via data-centric alignment.

The rapid progress of large language models (LLMs) has seen them excel and frequently surpass human performance on standard benchmarks. This has enabled many downstream applications, such as LLM agents, to rely on their sophisticated reasoning to navigate complex task requirements. However, LLMs are known to unexpectedly falter in simple tasks and under seemingly straightforward circumstances - underscoring the need for better and more diverse evaluation setups to measure their true capabilities. To this end, we choose to study compositional and conditional reasoning, two cornerstones of human cognition, and introduce GroundCocoa - a lexically diverse benchmark connecting these reasoning skills to the real-world problem of flight booking. Our task involves aligning detailed user preferences with available flight options presented in a multiple-choice format. Results indicate a significant disparity in performance among current state-of-the-art LLMs with even the best performing model, GPT-4 Turbo, not exceeding 67% accuracy despite advanced prompting techniques.

We propose LogicVista, an evaluation benchmark that assesses the integrated logical reasoning capabilities of multimodal large language models (MLLMs) in Visual contexts. Recent advancements in MLLMs have demonstrated various fascinating abilities, from crafting poetry based on an image to performing mathematical reasoning. However, there is still a lack of systematic evaluation of MLLMs' proficiency in logical reasoning tasks, which are essential for activities like navigation and puzzle-solving. Thus we evaluate general logical cognition abilities across 5 logical reasoning tasks encompassing 9 different capabilities, using a sample of 448 multiple-choice questions. Each question is annotated with the correct answer and the human-written reasoning behind the selection, enabling both open-ended and multiple-choice evaluation. A total of 8 MLLMs are comprehensively evaluated using LogicVista. Code and Data Available at https://github.com/Yijia-Xiao/LogicVista.

Recent proprietary large language models (LLMs), such as GPT-4, have achieved a milestone in tackling diverse challenges in the biomedical domain, ranging from multiple-choice questions to long-form generations. To address challenges that still cannot be handled with the encoded knowledge of LLMs, various retrieval-augmented generation (RAG) methods have been developed by searching documents from the knowledge corpus and appending them unconditionally or selectively to the input of LLMs for generation. However, when applying existing methods to different domain-specific problems, poor generalization becomes apparent, leading to fetching incorrect documents or making inaccurate judgments. In this paper, we introduce Self-BioRAG, a framework reliable for biomedical text that specializes in generating explanations, retrieving domain-specific documents, and self-reflecting generated responses. We utilize 84k filtered biomedical instruction sets to train Self-BioRAG that can assess its generated explanations with customized reflective tokens. Our work proves that domain-specific components, such as a retriever, domain-related document corpus, and instruction sets are necessary for adhering to domain-related instructions. Using three major medical question-answering benchmark datasets, experimental results of Self-BioRAG demonstrate significant performance gains by achieving a 7.2% absolute improvement on average over the state-of-the-art open-foundation model with a parameter size of 7B or less. Overall, we analyze that Self-BioRAG finds the clues in the question, retrieves relevant documents if needed, and understands how to answer with information from retrieved documents and encoded knowledge as a medical expert does. We release our data and code for training our framework components and model weights (7B and 13B) to enhance capabilities in biomedical and clinical domains.

Recent reasoning models through test-time scaling have demonstrated that long chain-of-thoughts can unlock substantial performance boosts in hard reasoning tasks such as math and code. However, the benefit of such long thoughts for system-2 reasoning is relatively less explored in other domains such as perceptual tasks where shallower, system-1 reasoning seems sufficient. In this paper, we introduce LongPerceptualThoughts, a new synthetic dataset with 30K long-thought traces for perceptual tasks. The key challenges in synthesizing elaborate reasoning thoughts for perceptual tasks are that off-the-shelf models are not yet equipped with such thinking behavior and that it is not straightforward to build a reliable process verifier for perceptual tasks. Thus, we propose a novel three-stage data synthesis framework that first synthesizes verifiable multiple-choice questions from dense image descriptions, then extracts simple CoTs from VLMs for those verifiable problems, and finally expands those simple thoughts to elaborate long thoughts via frontier reasoning models. In controlled experiments with a strong instruction-tuned 7B model, we demonstrate notable improvements over existing visual reasoning data-generation methods. Our model, trained on the generated dataset, achieves an average +3.4 points improvement over 5 vision-centric benchmarks, including +11.8 points on V^* Bench. Notably, despite being tuned for vision tasks, it also improves performance on the text reasoning benchmark, MMLU-Pro, by +2 points.

Recent works on large language models (LLMs) have successfully demonstrated the emergence of reasoning capabilities via reinforcement learning (RL). Although recent efforts leverage group relative policy optimization (GRPO) for MLLMs post-training, they constantly explore one specific aspect, such as grounding tasks, math problems, or chart analysis. There are no works that can leverage multi-source MLLM tasks for stable reinforcement learning. In this work, we present a unified perspective to solve this problem. We present Mixed-R1, a unified yet straightforward framework that contains a mixed reward function design (Mixed-Reward) and a mixed post-training dataset (Mixed-45K). We first design a data engine to select high-quality examples to build the Mixed-45K post-training dataset. Then, we present a Mixed-Reward design, which contains various reward functions for various MLLM tasks. In particular, it has four different reward functions: matching reward for binary answer or multiple-choice problems, chart reward for chart-aware datasets, IoU reward for grounding problems, and open-ended reward for long-form text responses such as caption datasets. To handle the various long-form text content, we propose a new open-ended reward named Bidirectional Max-Average Similarity (BMAS) by leveraging tokenizer embedding matching between the generated response and the ground truth. Extensive experiments show the effectiveness of our proposed method on various MLLMs, including Qwen2.5-VL and Intern-VL on various sizes. Our dataset and model are available at https://github.com/xushilin1/mixed-r1.

Recent progress in large language models (LLMs) has showcased impressive proficiency in numerous Arabic natural language processing (NLP) applications. Nevertheless, their effectiveness in Arabic medical NLP domains has received limited investigation. This research examines the degree to which state-of-the-art LLMs demonstrate and articulate healthcare knowledge in Arabic, assessing their capabilities across a varied array of Arabic medical tasks. We benchmark several LLMs using a medical dataset proposed in the Arabic NLP AraHealthQA challenge in MedArabiQ2025 track. Various base LLMs were assessed on their ability to accurately provide correct answers from existing choices in multiple-choice questions (MCQs) and fill-in-the-blank scenarios. Additionally, we evaluated the capacity of LLMs in answering open-ended questions aligned with expert answers. Our results reveal significant variations in correct answer prediction accuracy and low variations in semantic alignment of generated answers, highlighting both the potential and limitations of current LLMs in Arabic clinical contexts. Our analysis shows that for MCQs task, the proposed majority voting solution, leveraging three base models (Gemini Flash 2.5, Gemini Pro 2.5, and GPT o3), outperforms others, achieving up to 77% accuracy and securing first place overall in the Arahealthqa 2025 shared task-track 2 (sub-task 1) challenge. Moreover, for the open-ended questions task, several LLMs were able to demonstrate excellent performance in terms of semantic alignment and achieve a maximum BERTScore of 86.44%.

Vision-Language Models (VLMs) have demonstrated impressive world knowledge across a wide range of tasks, making them promising candidates for embodied reasoning applications. However, existing benchmarks primarily evaluate the embodied reasoning ability of VLMs through multiple-choice questions based on image annotations -- for example, selecting which trajectory better describes an event in the image. In this work, we introduce the Point-It-Out (PIO) benchmark, a novel benchmark designed to systematically assess the embodied reasoning abilities of VLMs through precise visual grounding. We propose a hierarchical evaluation protocol spanning three stages (S1: referred-object localization, S2: task-driven pointing, and S3: visual trace prediction), with data collected from critical domains for embodied intelligence, including indoor, kitchen, driving, and robotic manipulation scenarios. Extensive experiments with over ten state-of-the-art VLMs reveal several interesting findings. For example, strong general-purpose models such as GPT-4o, while excelling on many benchmarks (e.g., language, perception, and reasoning), underperform compared to some open-source models in precise visual grounding; models such as MoLMO perform well in S1 and S2 but struggle in S3, where requires grounding combined with visual trace planning.

Logical reasoning task involves diverse types of complex reasoning over text, based on the form of multiple-choice question answering. Given the context, question and a set of options as the input, previous methods achieve superior performances on the full-data setting. However, the current benchmark dataset has the ideal assumption that the reasoning type distribution on the train split is close to the test split, which is inconsistent with many real application scenarios. To address it, there remain two problems to be studied: (1) How is the zero-shot capability of the models (train on seen types and test on unseen types)? (2) How to enhance the perception of reasoning types for the models? For problem 1, we propose a new benchmark for generalized zero-shot logical reasoning, named ZsLR. It includes six splits based on the three type sampling strategies. For problem 2, a type-aware model TaCo is proposed. It utilizes both the heuristic input reconstruction and the contrastive learning to improve the type perception in the global representation. Extensive experiments on both the zero-shot and full-data settings prove the superiority of TaCo over the state-of-the-art methods. Also, we experiment and verify the generalization capability of TaCo on other logical reasoning dataset.

Metaphorical comprehension in images remains a critical challenge for Nowadays AI systems. While Multimodal Large Language Models (MLLMs) excel at basic Visual Question Answering (VQA), they consistently struggle to grasp the nuanced cultural, emotional, and contextual implications embedded in visual content. This difficulty stems from the task's demand for sophisticated multi-hop reasoning, cultural context, and Theory of Mind (ToM) capabilities, which current models lack. To fill this gap, we propose MetaphorStar, the first end-to-end visual reinforcement learning (RL) framework for image implication tasks. Our framework includes three core components: the fine-grained dataset TFQ-Data, the visual RL method TFQ-GRPO, and the well-structured benchmark TFQ-Bench. Our fully open-source MetaphorStar family, trained using TFQ-GRPO on TFQ-Data, significantly improves performance by an average of 82.6% on the image implication benchmarks. Compared with 20+ mainstream MLLMs, MetaphorStar-32B achieves state-of-the-art (SOTA) on Multiple-Choice Question and Open-Style Question, significantly outperforms the top closed-source model Gemini-3.0-pro on True-False Question. Crucially, our experiments reveal that learning image implication tasks improves the general understanding ability, especially the complex visual reasoning ability. We further provide a systematic analysis of model parameter scaling, training data scaling, and the impact of different model architectures and training strategies, demonstrating the broad applicability of our method. We open-sourced all model weights, datasets, and method code at https://metaphorstar.github.io.

Reinforcement learning from verifiable rewards (RLVR) has improved the reasoning abilities of LLMs, yet a fundamental limitation remains: models cannot learn from problems that are too difficult to solve under their current policy, as these yield no meaningful reward signal. We propose a simple yet effective solution based on task reformulation. We transform challenging open-ended problems into cognitively simpler variants -- such as multiple-choice and cloze formats -- that preserve the original answer while reducing the effective search space and providing denser learning signals. These reformulations span a spectrum from discriminative to generative tasks, which we exploit to bootstrap learning: models first learn from structured, easier formats, and this knowledge transfers back to improve performance on the original open-ended problems. Building on this insight, we introduce Cog-DRIFT, a framework that constructs reformulated variants and organizes them into an adaptive curriculum based on difficulty. Training progresses from easier to harder formats, enabling the model to learn from problems that previously yielded zero signal under standard RL post-training. Cog-DRIFT not only improves on the originally unsolvable hard problems (absolute +10.11% for Qwen and +8.64% for Llama) but also generalizes well to other held-out datasets. Across 2 models and 6 reasoning benchmarks, our method consistently outperforms standard GRPO and strong guided-exploration baselines. On average, Cog-DRIFT shows +4.72% (Qwen) and +3.23% (Llama) improvements over the second-best baseline. We further show that Cog-DRIFT improves pass@k at test time, and the curriculum improves sample efficiency. Overall, our results highlight task reformulation and curriculum learning as an effective paradigm for overcoming the exploration barrier in LLM post-training.

Recent deployments of large language models (LLMs) as autonomous trading agents raise questions about whether financial decision-making competence generalizes beyond specific market patterns and how it should be trained and evaluated in noisy markets lacking ground truth. We propose a structured framework for training and evaluating such models. Central to our approach is a curated, multiple-choice question (MCQ) dataset derived from classic textbooks and historical markets, verified by an AI committee, enriched with structured reasoning traces, and augmented to reduce shortcut learning. To evaluate whether performance on isolated MCQs generalizes to real-world trading, we introduce a two-stage protocol combining test-set evaluation with an MCQ-based chronological trading simulation. Extensive evaluations across market regimes provide statistically robust evidence that open models trained with our framework exhibit competitive, risk-aware behavior over time, outperform open-source baselines, and approach frontier-model performance at smaller scale. We release the dataset and evaluation framework to support further research.

Dental image analysis plays a pivotal role in supporting accurate diagnosis and treatment planning in oral healthcare. Although recent advances have produced dental AI models for specific tasks and individual imaging modalities, their isolated designs limit practical use in real-world clinical workflows. In this paper, we present OralAgent, the first dental-specialized AI agent that unifies multimodal reasoning, tool-based decision-making, and knowledge-grounded retrieval within an end-to-end automated framework. It integrates 22 visual analysis tools and 368 widely-used classical dental textbooks, enabling autonomous reasoning, planning, tool use, knowledge retrieval, and multi-step workflow execution. Furthermore, we introduce OralCorpus, a large-scale, high-quality bilingual textual resource containing 134.8M tokens curated for dental retrieval-augmented generation (RAG). To evaluate models' multidisciplinary dental knowledge, we construct OralQA-ZH, a Chinese multiple-choice question benchmark consisting of 798 items across eleven oral subspecialties. Extensive experiments demonstrate that OralAgent achieves state-of-the-art performance on the MMOral-Uni, MMOral-OPG, and OralQA-ZH benchmarks, highlighting its effectiveness, interpretability, and adaptability in real-world clinical settings. The code and models are publicly available at https://github.com/isjinghao/OralAgent.

The handling of probabilities in the form of uncertainty or partial information is an essential task for LLMs in many settings and applications. A common approach to evaluate an LLM's probabilistic reasoning capabilities is to assess its ability to answer questions pertaining to probability through the use of multiple-choice questions (MCQs). However, this paradigm, which we refer to as explicit probabilistic reasoning, has been shown in the literature to yield significant limitations (e.g., sensitivity to answer ordering). In this work, we introduce an alternative approach, named implicit probabilistic reasoning, which evaluates the models' ability to integrate probabilistic reasoning into their text generation process. To achieve this, we rephrase MCQs as text-completion scenarios with a determined set of outcomes and compare the model's next-token probability assignments to the true likelihood of the outcomes. In line with previous work, we find that models exhibit solid performance in their explicit probabilistic reasoning (i.e., answers to MCQs). However, during text completion (i.e., implicit probabilistic reasoning), where the same information must be taken into account to generate text, the models' predictions often significantly diverge from the known ground truth. For instance, our evaluation method reveals that implicit probabilistic reasoning is improperly influenced by many factors, such as independent prior events, partial observations about a result, or statistical background information. All of these issues can cause erroneous results to be produced in text generation, which are not detected by conventional MCQ-based evaluation.

Exam-style accuracy does not establish whether large language models (LLMs) reason well over clinical records. We define clinical reasoning as integrating and updating evidence across time and sources to form, revise and justify a patient's problem representation and a defensible plan. This structured narrative review maps three literatures: medical education assessment instruments, clinical LLM benchmarks published from 2023 onwards, and general-domain methods for evaluating long-form generation. We examine six dimensions: problem representation, temporal synthesis, differential and management reasoning, counterfactual reasoning, calibrated uncertainty, and reasoning faithfulness. Preprints are included and flagged. No single instrument covers all six dimensions. Problem representation and differential or management reasoning are reasonably covered, although reliability varies by instrument and setting. TIMER-Eval targets temporal synthesis, and ER-Reason assesses sequential diagnostic belief updating. Dedicated uncertainty and counterfactual evaluations are emerging, but their applicability to longitudinal free-text reasoning remains limited. Factual completeness is well theorised in general-domain evaluation, with early clinical evidence of important omissions. Faithfulness remains the weakest dimension, with one identified clinical causal-ablation study on multiple-choice questions. Existing tools should be combined through binary rubric items, separate completeness and correctness scores, case-specific importance weighting with non-compensable safety caps, temporal order-consistency checks, and chance-corrected reliability reporting. Further design work is needed for calibrated uncertainty, counterfactual reasoning and faithfulness over longitudinal free-text records. This review provides a design rationale, not a validated instrument.

We present the ARC-DA dataset, a direct-answer ("open response", "freeform") version of the ARC (AI2 Reasoning Challenge) multiple-choice dataset. While ARC has been influential in the community, its multiple-choice format is unrepresentative of real-world questions, and multiple choice formats can be particularly susceptible to artifacts. The ARC-DA dataset addresses these concerns by converting questions to direct-answer format using a combination of crowdsourcing and expert review. The resulting dataset contains 2985 questions with a total of 8436 valid answers (questions typically have more than one valid answer). ARC-DA is one of the first DA datasets of natural questions that often require reasoning, and where appropriate question decompositions are not evident from the questions themselves. We describe the conversion approach taken, appropriate evaluation metrics, and several strong models. Although high, the best scores (81% GENIE, 61.4% F1, 63.2% ROUGE-L) still leave considerable room for improvement. In addition, the dataset provides a natural setting for new research on explanation, as many questions require reasoning to construct answers. We hope the dataset spurs further advances in complex question-answering by the community. ARC-DA is available at https://allenai.org/data/arc-da

Multimodal Large Language Models (MLLMs) have achieved significant advancements in tasks like Visual Question Answering (VQA) by leveraging foundational Large Language Models (LLMs). However, their abilities in specific areas such as temporal understanding, which is crucial for comprehending real-world dynamics, remain underexplored. To address this, we propose a challenging evaluation benchmark named TemporalVQA, consisting of two parts: (1) Temporal Order Understanding and (2) Time-lapse Estimation. The first part requires MLLMs to determine the sequence of events by analyzing temporally consecutive video frames. The second part presents image pairs with varying time differences, framed as multiple-choice questions, asking MLLMs to estimate the time-lapse between images with options ranging from seconds to years. Our evaluations of advanced MLLMs, including models like GPT-4o and Gemini-1.5-Pro, reveal significant challenges: GPT-4o achieved only 43.8% average consistent accuracy in temporal order tasks and 70% in time-lapse estimation, with open-source models performing even less effectively. These findings underscore the limitations of current MLLMs in visual temporal understanding and reasoning, highlighting the need for further improvements in their temporal capabilities. Our dataset can be found at https://huggingface.co/datasets/fazliimam/temporal-vqa.

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

Recent work shows that reinforcement learning(RL) can markedly sharpen the reasoning ability of large language models (LLMs) by prompting them to "think before answering." Yet whether and how these gains transfer to audio-language reasoning remains largely unexplored. We extend the Group-Relative Policy Optimization (GRPO) framework from DeepSeek-R1 to a Large Audio-Language Model (LALM), and construct a 32k sample multiple-choice corpus. Using a two-stage regimen supervised fine-tuning on structured and unstructured chains-of-thought, followed by curriculum-guided GRPO, we systematically compare implicit vs. explicit, and structured vs. free form reasoning under identical architectures. Our structured audio reasoning model, SARI (Structured Audio Reasoning via Curriculum-Guided Reinforcement Learning), achieves a 16.35% improvement in average accuracy over the base model Qwen2-Audio-7B-Instruct. Furthermore, the variant built upon Qwen2.5-Omni reaches state-of-the-art performance of 67.08% on the MMAU test-mini benchmark. Ablation experiments show that on the base model we use: (i) SFT warm-up is important for stable RL training, (ii) structured chains yield more robust generalization than unstructured ones, and (iii) easy-to-hard curricula accelerate convergence and improve final performance. These findings demonstrate that explicit, structured reasoning and curriculum learning substantially enhances audio-language understanding.

In this paper, we introduce BMMR, a large-scale bilingual, multimodal, multi-disciplinary reasoning dataset for the community to develop and evaluate large multimodal models (LMMs). BMMR comprises 110k college-level questions spanning 300 UNESCO-defined subjects, spanning diverse formats-multiple-choice, fill-in-the-blank, and open-ended QA-and sourced from both print and digital media such as books, exams, and quizzes. All data are curated and filtered via a human-in-the-loop and scalable framework, and each instance is paired with a high-quality reasoning path. The dataset is organized into two parts: BMMR-Eval that comprises 20,458 high-quality instances to comprehensively assess LMMs' knowledge and reasoning across multiple disciplines in both Chinese and English; and BMMR-Train that contains 88,991 instances to support further research and development, extending the current focus on mathematical reasoning to diverse disciplines and domains. In addition, we propose the process-based multi-discipline verifier (i.e., BMMR-Verifier) for accurate and fine-grained evaluation of reasoning paths. Extensive experiments on 24 models reveal that (i) even SOTA models (e.g., o3 and Gemini-2.5-Pro) leave substantial headroom on BMMR-Eval; (ii) reasoning models exhibit discipline bias and outperform LMMs only on specific subjects; (iii) open-source models still trail their proprietary counterparts; and (iv) fine-tuning on BMMR-Train narrows this gap. Additionally, we conduct reasoning-chain analyses using BMMR-Verifier and other in-depth studies, uncovering the challenges LMMs currently face in multidisciplinary reasoning. We will release the data, and we hope our work can offer insights and contributions to the community.

Visual transformation reasoning (VTR) is a vital cognitive capability that empowers intelligent agents to understand dynamic scenes, model causal relationships, and predict future states, and thereby guiding actions and laying the foundation for advanced intelligent systems. However, existing benchmarks suffer from a sim-to-real gap, limited task complexity, and incomplete reasoning coverage, limiting their practical use in real-world scenarios. To address these limitations, we introduce VisualTrans, the first comprehensive benchmark specifically designed for VTR in real-world human-object interaction scenarios. VisualTrans encompasses 12 semantically diverse manipulation tasks and systematically evaluates three essential reasoning dimensions - spatial, procedural, and quantitative - through 6 well-defined subtask types. The benchmark features 472 high-quality question-answer pairs in various formats, including multiple-choice, open-ended counting, and target enumeration. We introduce a scalable data construction pipeline built upon first-person manipulation videos, which integrates task selection, image pair extraction, automated metadata annotation with large multimodal models, and structured question generation. Human verification ensures the final benchmark is both high-quality and interpretable. Evaluations of various state-of-the-art vision-language models show strong performance in static spatial tasks. However, they reveal notable shortcomings in dynamic, multi-step reasoning scenarios, particularly in areas like intermediate state recognition and transformation sequence planning. These findings highlight fundamental weaknesses in temporal modeling and causal reasoning, providing clear directions for future research aimed at developing more capable and generalizable VTR systems. The dataset and code are available at https://github.com/WangYipu2002/VisualTrans.

We introduce the Korean Canonical Legal Benchmark (KCL), a benchmark designed to assess language models' legal reasoning capabilities independently of domain-specific knowledge. KCL provides question-level supporting precedents, enabling a more faithful disentanglement of reasoning ability from parameterized knowledge. KCL consists of two components: (1) KCL-MCQA, multiple-choice problems of 283 questions with 1,103 aligned precedents, and (2) KCL-Essay, open-ended generation problems of 169 questions with 550 aligned precedents and 2,739 instance-level rubrics for automated evaluation. Our systematic evaluation of 30+ models shows large remaining gaps, particularly in KCL-Essay, and that reasoning-specialized models consistently outperform their general-purpose counterparts. We release all resources, including the benchmark dataset and evaluation code, at https://github.com/lbox-kr/kcl.