QA
-
Selective State-Space Adaptation and Retrieval for Language Model Reasoning
Abstract
Low-rank adaptation introduces a static learned update applied identically to every input. The update provides task-level adaptation but does not explicitly represent token-level or instance-level state variation. A family of adapters is proposed that introduces selective state-space recurrence at two complementary granularities. At the token level, \textbf{MaLoRA} (Mamba-modulated low-rank adaptation) makes the adapter's scaling factor a dynamic input-dependent function with recurrent state across tokens, in contrast to the stateless modulators of prior work. At the context level, \textbf{MaRA} (Mamba Retrieval Adapter) tracks cross-segment state and selects the segments most relevant to the query, before the modulated language model generates its answer. Across three frozen backbones (Qwen-2.5-7B, Llama-3.1-8B, Gemma-2-9B) and two reasoning benchmarks (MuSiQue, 2WikiMultihopQA), the family improves reasoning accuracy on every cell of the $3{\times}2$ grid, by $+6.8$ F1 ($+10.5\%$ relative) on average and up to $+9.3$ F1 ($+18.2\%$ relative) on the hardest cell over the LoRA baseline, and the token-level gains carry to RULER QA-2 under length stress.
Abstract (translated)
URL
https://arxiv.org/abs/2607.19326
PDF
-
MIRA-Ev:A Benchmark for Granular Evidence Detection and Relational Reasoning in Clinical Exams
2026-07-21 15:34:44Iker De la Iglesia, Johanna Ramirez-Romero, Jose Maria Villa-Gonzalez, Irune Urroz Garc\'ia, Ander Barrena, Aitziber AtutxaAbstract
Clinical NLP evaluation remains dominated by multiple-choice question answering (MCQA), which scores only final-answer accuracy and cannot detect when a model reaches the correct diagnosis while grounding it in irrelevant, absent, or contradictory evidence. We introduce MIRA-Ev, a clinical argument mining benchmark built on Spanish Médico Interno Residente (MIR) licensing-exam cases, re-annotated by expert clinicians with span-level premises, claims, and directed support/attack relations, and released in parallel Spanish (native), English, and Basque versions, the first clinical argumentation resource in Basque. MIRA-Ev organizes evaluation into a three-tier task hierarchy: evidence sentence retrieval, argumentative component extraction, and relation classification.
Abstract (translated)
URL
https://arxiv.org/abs/2607.19201
PDF
-
DAIS: Dependency-Aware Intermediate QA Supervision for Complex Reasoning
2026-07-21 13:25:00Yu Wang, Ming Fan, Xicheng Zhang, Zhiyong Li, Zhihu Wang, Caiyue Xu, Dahai Hu, Ting LiuAbstract
Chain-of-thought (CoT) supervision exposes intermediate rationales, but flat rationale targets usually optimize a single reasoning sequence and provide limited supervision on how local conclusions should support later decisions. We introduce Dependency-Aware Intermediate QA Supervision (DAIS), a training-time framework that converts filtered teacher rationales into stage-level QA records. Each intermediate record predicts a local answer conditioned on the previous states needed for that decision, while the final-answer record keeps the original task format; evaluation therefore uses only the original input and optional context. Across GDPR, AIACT, MedQA, and FOLIO with multiple Qwen backbones, DAIS improves average final-answer accuracy over answer-only, flat chain-of-thought, and independent-QA baselines. On policy-compliance benchmarks, it achieves a largest gain of 5.6% and an average gain of 4.2% over the strongest non-DAIS baseline. Controlled ablations show that valid previous-state conditioning contributes beyond longer targets or additional intermediate text, supporting dependency-conditioned intermediate QA as a lightweight auxiliary supervision signal for standard final-answer inference.
Abstract (translated)
URL
https://arxiv.org/abs/2607.19088
PDF
-
Quality Action Assurance: Multimodal Verification of Examiner Claims in VR OSCEs
2026-07-21 12:51:42Harry Rogers, Sally Shiels, Ashley Tomlinson, James Thomas, James Aylward, Nathan Gauge, Helen Higham, Alison NobleAbstract
Objective Structured Clinical Examinations (OSCEs) are the gold standard for assessing clinical competence, yet scoring remains vulnerable to examiner subjectivity, fatigue, and cognitive bias. Standard examiner validation via inter-rater statistics lacks explanatory power regarding the source of errors, as it neither analyzes examiner reasoning nor verifies examiner claims against actual events. Thus, we introduce Quality Action Assurance (QAA), a multimodal framework that verifies examiner claims in Virtual Reality (VR) pediatric OSCEs by comparing actions claimed by examiners against the true sequence of events, constructed from video, VR logs, and actor data. QAA combines a constrained temporal action alignment model, which performs action localization and actor source attribution, with a large language model that extracts examiner claims and checks them against the record. Across a 5-fold cross-validation, QAA achieves 99.2% $\pm$ 0.7% Actor F1 and 93.4% $\pm$ 1.9% W@16 for temporal alignment. Overall, QAA detects examiner errors with 70.0% precision and 76.7% recall, improving factual correctness from 39.2% to 79.2%, enabling fairer OSCE assessment.
Abstract (translated)
URL
https://arxiv.org/abs/2607.19063
PDF
-
Dual Adversarial Fine-tuning for Enhancing Robustness of Large Vision Language Model
2026-07-21 10:49:29Sibo Wang, Jie Zhang, Shiguang Shan, Xilin Chen, Wen GaoAbstract
While Large Vision-Language Models (LVLMs), represented by LLaVA and GPT-4V, have demonstrated remarkable capabilities, their visual inputs remain vulnerable to adversarial attacks, posing significant security risks. Existing defense methods predominantly target single-task scenarios (e.g., zero-shot classification) and consequently lack generalizability across various multimodal tasks. To address this limitation, we propose a dual adversarial fine-tuning framework that jointly optimizes visual and semantic supervision signals from two modalities, enhancing model robustness while generalizing across multiple downstream tasks. The proposed framework comprises two core components, i.e., $\textbf{Visual}$ supervision branch and $\textbf{Semantic}$ supervision branch. The former branch leverages features from clean images, extracted via a frozen original vision encoder, to guide adversarial robustness while the latter incorporates caption-image alignment as a contextual signal to preserve semantic coherence under attack. Moreover, our method achieves cross-task robustness by simply replacing the CLIP vision encoder in the original model, with no need of separate task-specific retraining or architecture this http URL experiments demonstrate that our approach outperforms the state-of-the-art method in adversarial robustness evaluation across zero-shot classification, image captioning, and visual question answering (VQA) tasks.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18958
PDF
-
TAP-RAG: Task-Aware Policy Control for Long-Document Multimodal Question Answering
2026-07-21 09:59:13Zhong Ji, Keqi Jin, Yan Zhang, Jiasheng LiAbstract
Long-document multimodal question answering requires more than retrieving relevant chunks from a large document. Different queries require different evidence behavior. Existing multimodal RAG systems improve evidence access through text chunks, page images, graph links, or heterogeneous document elements, but they often apply a largely query-agnostic evidence-use strategy. We present TAP-RAG, a task-aware policy-controlled RAG framework for long-document multimodal QA. TAP-RAG contains a main controller, the Task-Aware Policy Controller (TAPC), and two policy-guided evidence executors: Task-Aware Query-Guided Flow Diffusion (TA-QFD) and Task-Aware Visual Enhancement (TAVE). For each query, TAPC predicts the task prior, estimates visual/local/global evidence signals, and produces an executable policy. TA-QFD then expands textual and structural evidence over the multimodal document graph, while TAVE selectively inspects page images when visual or layout evidence is needed. A guarded synthesis stage fuses text, visual, and structural evidence and abstains when support is insufficient. On DocBench and MMLongBench-Doc, TAP-RAG achieves the best overall accuracy among the compared systems, improving over a matched multimodal-RAG baseline by +9.1 points (61.1 to 70.2) and +4.5 points (42.2 to 46.7), respectively.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18917
PDF
-
OPD-IAD: From Language Judgment to Industrial Anomaly Detection via On-Policy Self-Distillation
2026-07-21 08:37:01Shuimu Chen, Jing Jin, Nan Su, Hongbo Xu, Zebang Cheng, Wenming Yang, Fei Ma, Guijin WangAbstract
Large vision-language models (LVLMs) have recently shown strong potential for industrial anomaly detection (IAD) by providing image-level anomaly judgments and interpretable defect reasoning. However, current LVLM-based IAD methods still struggle to produce precise pixel-level anomaly maps from generated language judgments. We aim to achieve precise pixel-level localization while using language as guidance rather than letting it dominate the visual response. Specifically, we propose \textbf{OPD-IAD}, an evidence-privileged dense on-policy self-distillation framework for LVLM-based IAD. OPD-IAD distills privileged defect evidence onto the model's own on-policy judgment trajectory, enabling the final generated judgment to be learned under dense supervision rather than treated only as a textual answer. The resulting judgment serves as a semantic condition for dense anomaly perception. To turn this condition into dense visual evidence, we introduce \textbf{Language-guided Visual Anchoring}, which uses a judgment reforward to re-encode the image and question under the final-judgment condition into semantic anchors and contrasts them with dense visual features through a contrastive heatmap head to generate anomaly maps. The language judgment therefore provides compact semantic guidance, while dense visual features remain the basis for pixel-level scoring, allowing language to guide anomaly localization without letting language quality directly dictate the pixel-level response. Extensive experiments show that OPD-IAD achieves the best overall performance among LVLM-based IAD methods, leading on most image-level, pixel-level, and QA metrics.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18850
PDF
-
Evaluating medical AI under missing information: same-provider judges and human raters change apparent safety
Abstract
Readiness stress-testing of medical AI has focused on closed-ended and multimodal benchmarks. We extend it to open-ended clinical conversation under missing information, where safe behavior means recognizing absent information and qualifying, clarifying, or not over-committing - and where the evaluator becomes part of the measurement. We stress-test four models - three flagships (Claude Opus 4.8, GPT-5.5, Grok 4.3) and one mid-tier model (Gemini 3.5 Flash) - by deleting the latter half of the final user turn in HealthBench conversations, grading responses with a four-provider LLM-judge panel and a blinded clinician-anchored reference. Two evaluator-facing results are robust. First, judge choice materially changes apparent safety: inter-judge agreement is only moderate (Fleiss' kappa = 0.65), and after adjusting for each judge's general leniency (vote-level logistic regression), a positive same-provider association remains (exact permutation p = 0.04; GPT-5.5 ~ +0.10 on the probability scale) - large enough to change which model appears to over-commit least once its own-provider judge is excluded. Second, LLM judges are more permissive than clinicians on a blinded 50-item subsample: all four are significantly more lenient than the stricter independent clinician (crediting appropriate uncertainty on 66-84% of items vs 52%), and three of four than the author-influenced consensus (Grok directional only; judge-vs-consensus kappa = 0.20-0.43). On the author-audited clinical-underdetermined subset the permissiveness gap widened and the point-estimate model ordering held. A closed-ended MedQA anchor confirms accuracy is high and option-order effects are within a +/-5-point equivalence region for three of four models, so the safety gap is about calibration, not knowledge. We release the harness, prompts, per-item outputs, judge panel, perturbation audit, and human-annotation protocol.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18828
PDF
-
AILQA: Evaluating AI-Driven Legal Question Answering Systems for the Indian Legal System
2026-07-21 08:01:17Shubham Kumar Nigam, Shubham Kumar Mishra, Noel Shallum, Kripabandhu Ghosh, Arnab BhattacharyaAbstract
This comprehensive study introduces an advanced Artificial Intelligence for Indian Legal Question Answering (AILQA) system tailored to the Indian legal context. AILQA leverages a variety of embedding and generative models, including recent Large Language Models (LLMs), to address the unique challenges posed by the intricate and diverse nature of Indian legal texts and to enhance the accuracy and reliability of responses to legal questions. We conducted rigorous evaluations using both lexical and semantic metrics, enriched by expert legal feedback, to ensure relevance and accuracy. Our findings underscore the effectiveness of the Retrieval-Augmented Generation (RAG) paradigm in improving answer quality, particularly in complex legal domains. Additionally, we assessed performance on standardized tests such as the All India Bar Examination (AIBE), thereby providing a robust benchmark for practical applications. Under the study's evaluation protocol, some AI-generated responses received higher ratings than the available reference answers, particularly when they contained accurate and relevant supporting details. This finding is specific to the evaluated dataset and rating criteria and should not be interpreted as evidence that the models generally outperform qualified legal professionals. We also discuss the challenges encountered, such as the need for precise context and the risks of model hallucination, and propose directions for future research to further refine AI capabilities in the legal field. This study aims to pave the way for enhanced legal decision-support systems, making them more accessible and effective for legal professionals and the public alike.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18825
PDF
-
Find Before You Fine-Tune: A Diagnostic Study of Small LLMs for Cybersecurity QA
2026-07-21 05:32:40Shaswata Mitra, Subash Neupane, Trisha Chakraborty, Himanshu Tripathi, Sudip Mittal, Aritran Piplai, Shahram RahimiAbstract
Large Language Models (LLMs) are increasingly fine-tuned for critical-domain Question-Answering (QA), yet choosing which small model to adapt, before paying the cost of adaptation, remains difficult. Fine-tuning can improve domain alignment, but it may also erode prior knowledge, weaken instruction-following, or increase hallucination, especially when labeled data are scarce or rapidly evolving as in cybersecurity. We present FiT (Find before Fine-Tune), a task-oriented diagnostic framework that characterizes small LLMs along three capabilities required for cybersecurity QA: vocabulary recognition, parametric knowledge, and contextualization of retrieved information. Using FiT, we conduct an empirical study of five open-weight 7-billion-parameter models under two fine-tuning regimes. We find that fine-tuning does not uniformly help: it consistently degrades vocabulary and parametric knowledge in small models, and the two regimes trade off differently. Knowledge-focused tuning causes moderate, rank-preserving degradation, whereas instruction-focused tuning collapses measured knowledge through induced abstention, inverting the knowledge ranking while leaving retrieval-grounded contextualization essentially intact. We quantify these regime-specific patterns with rank-correlation analysis and show that pre-fine-tuning FiT scores anticipate the direction of post-tuning change. Our results suggest that task-oriented diagnosis can screen out unsuitable models, avoid unnecessary fine-tuning, and support safer deployment of small LLMs in cybersecurity QA pipelines.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18725
PDF
-
Continual Video-MLLM Adaptation over Evolving Domains
2026-07-21 05:16:36Rui Cheng, Meixing Shi, Yuxiang Cai, Jingcai Guo, Jianwei Yin, Zhi ChenAbstract
Video multimodal large language models have shown strong capability in video understanding, yet their adaptation to sequentially evolving domains remains underexplored. In real-world deployments, video data often arrives continuously from heterogeneous domains, requiring the model to acquire new domain-specific knowledge without overwriting previously learned capabilities. Existing continual learning methods typically rely on shared adaptation spaces, which can induce severe cross-domain interference and catastrophic forgetting. We propose Distribution-Aware Expert Routing, a parameter-efficient framework for continual Video-MLLM adaptation over evolving domains. DAER maintains domain-isolated lightweight experts while keeping the pretrained Video-MLLM backbone frozen, thereby decoupling domain-specific adaptation from the general multimodal knowledge of the pretrained model. To enable fine-grained specialization, we introduce an intra-domain distribution-aware routing mechanism that matches each input to expert-level prototype reservoirs using MMD. To address the absence of task identities at inference time, we further propose an inter-domain routing mechanism that performs prototype matching in a discriminative subspace for robust domain identification. In addition, we introduce adaptive domain merging to improve parameter scalability and adopt a two-stage optimization strategy to stabilize expert specialization during continual learning. We evaluate DAER by curating a domain-incremental benchmark built from ten VidQA datasets covering diverse visual environments and reasoning demands. Experiments on two strong Video-MLLM backbones show that DAER consistently outperforms prior methods.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18716
PDF
-
RoboInter1.5: A Holistic Intermediate Representation Suite for Embodied World Modeling and Robotic Manipulation
2026-07-21 05:05:01Ziqin Wang, Hao Li, Weijun Wang, Junhao Cai, Jia Zeng, Yilun Chen, Jiangmiao Pang, Si LiuAbstract
Existing robot datasets remain expensive to curate, embodiment-specific, and insufficiently annotated with the fine-grained structure required for generalizable reasoning, execution, or long-horizon environment dynamics simulation. Building on our prior work, RoboInter1.0, we present RoboInter1.5, an extended and holistic suite of intermediate representations for both robotic manipulation and embodied world modeling. RoboInter1.5 provides a unified resource of data, benchmarks, and models centered on dense manipulation-oriented intermediate representations. Specifically, RoboInter-Data contains over 230k manipulation episodes across 571 scenes with dense per-frame annotations covering more than ten types of intermediate representations, including subtasks, primitive skills, object and gripper grounding, segmentation, affordance, grasp poses, contact points, motion traces, etc. Built upon these annotations, RoboInter-VQA introduces spatial and temporal embodied VQA tasks to benchmark and improve the intermediate-representation reasoning capabilities of our RoboInter-VLM. RoboInter-VLA further studies how such representations benefit action execution through implicit, explicit, and modular plan-then-execute paradigms. To better model the physical world, we further introduce RoboInter-World, which leverages intermediate representations as structured conditioning signals for controllable prediction of future world states. Extensive evaluations demonstrate that RoboInter1.5 provides a unified spatiotemporal scaffolding for intermediate representations. Rather than treating intermediate representations merely as interpretable signals, RoboInter1.5 conceptualizes them as a bidirectional interface that both regularizes low-level action spaces and constrains the latent rollouts of open-world physical simulators.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18709
PDF
-
Querying Multimodal Scientific Papers with AI: Practices and Preferences Across Blind, Low-Vision, and Sighted Scientists
2026-07-20 21:14:26Arnavi Chheda-Kothary, Lucy Lu Wang, Joseph Chee Chang, Jonathan BraggAbstract
Visual diagrams, figures, and tables are central to scientific papers, and convey information beyond what is captured in text. While blind or low-vision (BLV) scientists have traditionally relied on static alternative text to access figures in papers, the rise of artificial intelligence (AI) has made interactive question-answering (QA) a feasible paradigm for visual exploration; yet little is known about how scientists use visual QA in practice or how to improve its accessibility. In this work, we interview five BLV and five sighted scientists across different STEM fields to understand how they use two AI tools, ChatGPT and Gemini, to query multimodal scientific documents. Our findings characterize how scientists review multimodal content, including existing practices (along with accessibility workarounds) for engaging with visuals, and feedback on the suitability of AI-generated responses to multimodal queries. We further find that vague or incomplete image descriptions, as well as incorrect AI outputs more broadly, can cause both BLV and sighted scientists to abandon AI workflows. To support future research, we additionally contribute a dataset of 115 queries and responses from our participants' interactions with the AI tools for papers in their field. We close by discussing implications for AI-powered scientific QA systems, emphasizing considerations for access across abilities and domains.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18514
PDF
-
Search-on-Graph-R1: Training Large Language Models to Search Knowledge Graphs with Reinforcement Learning
2026-07-20 19:58:32Jia Ao Sun, Hao Yu, Fengran Mo, Zhan Su, Yuchen Hui, Bang Liu, Jian-Yun NieAbstract
Knowledge graph question answering (KGQA) requires navigating from topic entities to an answer several relations away. Recent methods prompt a frontier LLM to explore the graph through a retrieval tool, but their reliance on frontier-scale inference makes them costly to deploy. We present Search-on-Graph-R1 (\sogrone{}), which internalizes this navigation into a compact 8B model through supervised fine-tuning (SFT) followed by reinforcement learning (RL). Our central idea is to scaffold a frontier teacher with each question's gold SPARQL query, so the teacher traverses a known answer-bearing path with a live \texttt{Search} tool rather than having to discover the path itself. Since every call executes against a live Freebase server, the resulting trajectories are grounded in the knowledge graph by construction. On WebQSP, CWQ, and GrailQA, \sogrone{} at 8B surpasses every frozen frontier-LLM system in our comparison and posts the strongest results on CWQ of any system we compare against. It does so using no auxiliary module at inference and no LLM judge during training. Isolating each training stage shows that SFT and RL contribute complementary gains, our approach transfers across model families, and RL learns to reach answers in fewer \texttt{Search} calls than its SFT initialization.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18481
PDF
-
FinSAgent: Corpus-Aligned Multi-Agent RAG Framework for Evidence-Grounded SEC Filing Question Answering
2026-07-20 16:03:15Jijun Chi (University of Toronto), Zhenghan Tai (SimpleWay.AI, University of Toronto), Hanwei Wu (SimpleWay.AI, McMaster University), Tung Sum Thomas Kwok (SimpleWay.AI, University of California, Los Angeles), Hailin He (SimpleWay.AI), Zixing Liao (SimpleWay.AI), Bohuai Xiao (SimpleWay.AI), Chaolong Jiang (SimpleWay.AI), Jianliang Lei (SimpleWay.AI), Jerry Huang (Universit\'e de Montr\'eal, Mila - Quebec AI Institute), Peng Lu (Universit\'e de Montr\'eal), Muzhi Li (The Chinese University of Hong Kong), Liheng Ma (SimpleWay.AI, McGill University, Mila - Quebec AI Institute), Yihong Wu (Universit\'e de Montr\'eal), Sicheng Lyu (SimpleWay.AI, McGill University, Mila - Quebec AI Institute), Jingrui Tian (McGill University), Yihan Li (Boston University), Yanzhang Ma (SimpleWay.AI, Lakehead University), Dingtao Hu (McGill University), Yufei Cui (McGill University), Ling Zhou (CG Matrix Technology Limited), Lei Ding (SimpleWay.AI, University of Manitoba), Xinyu Wang (SimpleWay.AI, McGill University)Abstract
Financial question answering over U.S. Securities and Exchange Commission (SEC) filings requires retrieving and synthesizing heterogeneous evidence dispersed across long, standardized, and highly redundant disclosures. Existing retrieval-augmented and multi-agent systems typically derive retrieval queries directly from the user's question and rank candidates by semantic similarity. Together, these choices create prior-corpus misalignment: a mismatch between model priors and the target filings' structure, terminology, and evidence standards. As a result, query generation misses corpus-specific evidence, while semantic reranking favors topically similar but evidentially invalid false-positive chunks. We propose FinSAgent, an evidence-grounded multi-agent framework that reframes SEC filing QA as corpus-aligned retrieval planning and corrects both ends with a single principle: inject corpus-side conditioning wherever model priors would otherwise dominate. FinSAgent combines (1) role-specialized agents anchored to the mandated 10-K item structure, (2) database-aware query decomposition that conditions each agent's sub-queries on a lightweight, summary-level view of the local corpus, and (3) multi-path retrieval with a learned feature-gated reranker that separates evidential validity from semantic similarity. Across five offline financial QA benchmarks, FinSAgent improves retrieval coverage and answer correctness over strong single-agent and multi-agent baselines; in a three-arm randomized online experiment with 1,000 anonymous user ratings, it also receives higher scores than baselines.
Abstract (translated)
URL
https://arxiv.org/abs/2607.18102
PDF
-
Measuring and Improving Complex-Atomic Answer Consistency in Endoscopic VQA
Abstract
Endoscopic visual question answering (VQA) increasingly asks complex questions that combine several endoscopic answer components rather than isolated factual queries. Such complex answers may be scored as correct even when the same model fails on associated atomic questions. We introduce EndoCA, a paired complex-atomic answer consistency benchmark for evaluating whether complex answers remain consistent with same-image atomic answers. EndoCA contains two suites: EndoCA-Core evaluates compact question-complexity patterns commonly seen in practical endoscopic VQA, and EndoCA-Diagnostic supports controlled analysis across increasing question complexity. We evaluate 11 VLMs spanning open, medical, endoscopy-adapted, and closed-source models on EndoCA. Some VLMs achieve high complex-answer accuracy, yet their atomic-answer accuracy and complex-atomic answer consistency remain substantially lower. To reduce this complex-atomic inconsistency, we introduce Atomic-Support Reconciliation (ASR), a training-free mechanism that uses model-generated atomic answers as contextual premises for answer revision and consistency-guided selective answering. On four selected publicly available models, ASR-Revise improves paired complex-atomic correctness with modest changes in complex-answer accuracy, while ASR-Selective improves accuracy on answered cases by allowing the model to abstain from less reliable cases. Together, EndoCA and ASR provide a consistency-aware benchmark and a training-free mechanism for answer reconciliation and selective answering in endoscopic VQA.
Abstract (translated)
URL
https://arxiv.org/abs/2607.17834
PDF
-
Semantically Similar, Logically Distinct: Diagnosing the Semantic-Answerability Gap in Table RAG
2026-07-20 09:36:10Jiaming Tian, Liyao Li, Wentao Ye, Haobo Wang, Lihua Yu, Zujie Ren, Gang Chen, Junbo ZhaoAbstract
Tables are a critical knowledge source in retrieval-augmented generation (RAG), but a retrieved table may lack sufficient evidence to answer a query, a property we call answerability. While answerability broadly concerns whether a source or collection of sources contains sufficient evidence, retrieval models optimized for semantic relevance do not guarantee it even in the single-source case, creating a fundamental mismatch. To study this, we introduce TCR-Bench, a diagnostic benchmark for Table Content-level Answerability in RAG, built around sibling tables, i.e., tables with highly similar schemas but subtle content differences. On TCR-Bench, the dense retrievers we evaluate persistently exhibit a Semantic-Answerability Gap: they often retrieve the correct sibling group yet struggle to pinpoint the uniquely answerable table within it, dropping QA performance from 0.755 (oracle) to 0.330 (top-5 retrieved). Our analysis suggests this gap is associated with semantic accumulation, schema-level cue dependence, and weak row-column binding. As a diagnostic probe into the source of this gap, we test whether a lightweight two-stage pipeline, Answerability-Aware Reranking (AAR), applying direct query-table answerability judgment, can recover performance: it raises top-1 target retrieval from 18.2% to 57.4%, and this large gain is itself evidence that much of the observed failure reflects a missing answerability verification step, rather than an inherent limitation of model capacity alone.
Abstract (translated)
URL
https://arxiv.org/abs/2607.17742
PDF
-
Oracle Gap and Signal Fidelity: A Fixed-Pool Diagnostic for Test-Time Collaboration
Abstract
Test-time collaboration, including self-consistency, best-of-N selection, critic models, and verifier pipelines, is often credited with broadly improving LLM reasoning, yet its gains are uneven and sometimes negative. We ask when training-free collaboration should be expected to help. For a fixed candidate pool, we decompose a selector or verifier's net gain into measurable factors: recoverable mass, verification-signal coverage, conditional selection quality, and harm to already-correct outputs. This reframes collaboration as a candidate-selection problem rather than as an intrinsic property of a multi-agent topology. Across LiveCodeBench, MATH Level-5 hard subjects, and GPQA-Diamond, gains are bounded first by the oracle gap and then by signal fidelity, which we measure directly as candidate-level agreement between verifier verdicts and official labels. On LiveCodeBench, a public-test verifier (MCC 0.825) gains +8.14 percentage points (pp) over a first-sample baseline; a generated-test verifier (MCC 0.248) improves by +2.70pp and is not statistically distinguishable from an LLM selector, but operates at near-zero harm versus the selector's 4.69% harm rate. On MATH, a symbolic answer-equivalence selector beats self-consistency by +4.67pp, while LLM selectors are negative. On GPQA-Diamond, recoverable mass is only 3.03% and 87.54% of candidate pools are answer-identical; a weaker model's pools shrink both further, suggesting that oracle gap is a joint property of task, model, and sampling configuration. Our framework yields a practical pre-deployment diagnostic: estimate the oracle gap, then measure coverage, signal fidelity, and harm before investing in collaboration.
Abstract (translated)
URL
https://arxiv.org/abs/2607.17531
PDF
-
Thinking in Video: Can Video Generators Really Reason About the Real World?
2026-07-20 03:56:43Yongheng Zhang, Guang Yang, Ruihan Hou, Qiguang Chen, Ziang Liu, Xiaolong Liu, Manman Zhang, Yanchao Hao, Zheng Wei, Hao Wu, Libo Qin, Peishan Dai, Yinghui Li, Di Yin, Xing SunAbstract
Recent advances in world models and video generation have given rise to an emerging reasoning paradigm that leverages video generative models to simulate, predict, and reason about real-world dynamics. We redefine this paradigm as Thinking in Video, where video is not merely an output artifact but a medium for constructing, extending, and verifying causal thought. However, this promise remains unverified: convincing rollouts may reflect memorized appearances rather than causal understanding, while existing metrics separate perceptual fidelity from semantic logic. To evaluate whether video generators support such reasoning, we introduce the Causal-Generative Dual-Judge (CGDJ), auditing World Model Consistency from two perspectives. Explicit Causal Perception tests whether a generator reads a video scenario as a reasoning problem through spatio-temporal flattened visual question answering, while Implicit Generative Perception-Prediction Gap evaluates whether it renders the causal consequence as a consistent future video. Applying CGDJ to representative open- and closed-source generators reveals a clear Perception-Prediction Gap: open-source models produce plausible dynamics despite near-zero explicit causal perception, whereas advanced closed-source systems show stronger but still limited alignment between reasoning and generation. Further analysis exposes audio-visual misalignment, where models verbalize correct causal logic more reliably than they render it, challenging the "world simulator" narrative.
Abstract (translated)
URL
https://arxiv.org/abs/2607.17523
PDF
-
Evaluating LLMs When They Do Not Know the Answer: Statistical Evaluation of Mathematical Reasoning via Comparative Signals
2026-07-20 02:29:52Zihan Dong, Zhixian Zhang, Yang Zhou, Can Jin, Ruijia Wu, Linjun ZhangAbstract
Evaluating mathematical reasoning in LLMs is constrained by limited benchmark sizes and inherent model stochasticity, yielding high-variance accuracy estimates and unstable rankings across platforms. On difficult problems, an LLM may fail to produce a correct final answer, yet still provide reliable pairwise comparison signals indicating which of two candidate solutions is better. We leverage this observation to design a statistically efficient evaluation framework that combines standard labeled outcomes with pairwise comparison signals obtained by having models judge auxiliary reasoning chains. Treating these comparison signals as control variates, we develop a semiparametric estimator based on the efficient influence function (EIF) for the setting where auxiliary reasoning chains are observed. This yields a one-step estimator that achieves the semiparametric efficiency bound, guarantees strict variance reduction over naive sample averaging, and admits asymptotic normality for principled uncertainty quantification. Across simulations, our one-step estimator substantially improves ranking accuracy, with gains increasing as model output noise grows. Experiments on GPQA Diamond, AIME 2025, and GSM8K further demonstrate more precise performance estimation and more reliable model rankings, especially in small-sample regimes where conventional evaluation is pretty unstable.
Abstract (translated)
URL
https://arxiv.org/abs/2602.03061
PDF
