探索全球前沿学术脉络

01.

arXiv (CS.AI) 2026-06-12 DOI: arXiv:2606.12425

An Explainable AI Assistant for Introductory Programming Education: Improving Feedback Reliability with Instructor-AI Collaboration

作者:

Muntasir Hoq↗Griffin Pitts↗Bradford Mott↗Seung Lee↗Jessica Vandenberg↗Shuyin Jiao↗Narges Norouzi↗James Lester↗Bita Akram↗

arXiv:2606.12425v1 Announce Type: cross Abstract: Active learning is widely recognized as an effective approach for improving learning outcomes in introductory programming courses. However, insufficient instructional support often limits students' access to timely, personalized feedback, which is crucial for mastering foundational programming concepts. Although recent advances in AI, particularly large language models, offer scalable opportunities for feedback, concerns about explainability and reliability remain. In this paper, we present an AI-driven classroom assistant that leverages an explainable AI model to analyze student code, map logical errors to instructor-identified misconceptions, and deliver instructor-authored feedback, thereby grounding reliability in instructor-defined pedagogical knowledge. To evaluate the effectiveness of our framework, we conducted an expert evaluation to examine its alignment with instructor-verified feedback and deployed the system in a classroom setting to assess students' perceptions of its usability. Results indicate that the assistant can provide accurate, instructor-verified feedback to students while fostering a positive experience.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.CL) 2026-06-12 DOI: arXiv:2606.07218

HKVM-RAG: Key-Value-Separated Hypergraph Evidence Organization for Multi-Hop RAG

作者:

Mingyu Zhang↗Ying Ma↗

Multi-hop RAG poses a data-engineering problem beyond passage matching: under fixed retrieval budgets, a system must organize retrieved text into evidence units that expose answer chains. Dense retrievers score passages independently, while graph-based memories make associations explicit but often rely on pairwise or entity-centered keys that fragment multi-hop evidence. We present HKVM-RAG, a key-value-separated evidence-organization layer. It assembles answer-path hyperedges from cached passage-level LLM evidence tuples and uses them as retrieval keys, while retaining passage text as answer values. To isolate key-space design, our fixed-substrate protocol holds the tuple cache, candidate passages, reader, and evaluation budget constant across pairwise graph and hypergraph variants. Weighted hypergraph key-value retrieval improves over KG-PPR by +3.426 F1 on 2WikiMultiHopQA and +3.592 F1 on MuSiQue; HotpotQA shows that higher structured support coverage need not yield standalone answer-F1 gains. We therefore study WHG-KV as an evidence-control signal rather than a dense-retrieval replacement. Oracle and train-to-dev analyses identify support selection as repairable, and a dense-aware controller combines frozen ColBERTv2 and HKVM rank/score features using out-of-fold HKVM predictions. It reaches 88.846, 65.073, and 85.810 F1 on the three benchmarks, improving over ColBERTv2 by +11.084, +6.763, and +5.966 F1. Source-level ablations show that matched non-WHG structured signals do not match the WHG-KV gains. These results provide bounded evidence that key-value-separated hypergraph organization can serve as a reusable evidence-control mechanism for multi-hop RAG.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2606.17043

Hierarchical Advantage Weighting for Online RL Fine-Tuning of VLAs from Sparse Episode Outcomes

作者:

Tongyan Fang↗Siyuan Huang↗Naiyu Fang↗Ganlong Zhao↗Zhongjin Luo↗Jianbo Liu↗Xiaogang Wang↗Ying Dong↗Hongsheng Li↗

arXiv:2606.17043v1 Announce Type: cross Abstract: When pretrained VLA policies are fine-tuned through online RL, each rollout episode produces only a single binary outcome (success or failure), yet the actor update requires per-transition supervision. Existing approaches commonly reduce this sparse outcome to a single scalar reward or advantage signal, which conflates distinct forms of transition-level feedback and provides limited guidance once basic task success becomes achievable. First, a single scalar signal conflates the two objectives of viability and efficiency; once basic success is achieved, the binary label provides no gradient to distinguish efficient completions from slow ones. Second, real-world rollouts mix autonomous and intervention segments; naively assigning episode outcomes across these boundaries introduces incorrect credit assignment. To address these issues, we propose Hierarchical Advantage-Weighted Behavior Cloning (HABC), which trains separate critic heads for these two objectives on different data subsets and combines their outputs with a state-adaptive balance. A state-adaptive gate $g_t$ merges their one-step advantages, prioritizing viability when success is uncertain and shifting to efficiency only when viability is high, and converts the result into per-transition weights on the actor loss. Intervention-aware credit assignment further restricts outcome labels to segments executed by the current policy, preventing supervision from leaking across intervention boundaries. In real-robot experiments on three contact-rich bimanual tasks, HABC raises success from supervised fine-tuning (SFT) baselines of 36%, 44%, and 12% to 92%, 88%, and 38%.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2606.17477

Theoretical Grounding of Out-Of-Distribution Detection With Reinforcement Learning Optimizer

作者:

Salimeh Sekeh↗Xin Zhang↗

Out-of-distribution (OOD) detection in dynamic open-world environments requires a model to continually adapt to evolving data distributions while generalizing to covariate-shifted inputs and rejecting semantic-shifted OOD examples. Most existing OOD detection methods optimize only the current-step objective and do not explicitly account for how post-deployment environment changes affect future OOD behavior. In this paper, we establish a theoretical grounding for dynamic OOD detection using a reinforcement learning (RL)-guided optimizer that explicitly favors updates that reduce the semantic OOD false positive rate over time. We develop a novel augmented optimizer that uses an RL-guided correction term on top of standard gradient descent (GD) and show its improvement over both future-domain generalization and semantic-OOD rejection. We analyze temporal error decomposition in terms of model-change and environment-change generalization errors and develop a new theoretical framework for comparing the generalization errors under both GD and RL-guided optimizers.

阅读与讨论 → 访问原文 →

05.

medRxiv (Medicine) 2026-06-22 DOI: HASH:d82062e9eef9b4d7541979b8b723cf0b

Sex-specific multimorbidity clusters and all-cause mortality in relatively healthy older adults: findings from the ASPREE cohort

作者:

Du↗Vishwanath↗Wu↗Phyo↗A. Z. Z↗Orchard↗S. G↗Sheets↗Ernst↗M. E↗Ryan↗

Background: Multimorbidity is common in older adults, but sex differences in chronic condition clustering remain unclear. This study explored multimorbidity clusters and their associations with all-cause mortality among community-dwelling adults aged 70 years and over. Methods: This was a secondary analysis of data from 16,095 Australian ASPREE participants aged at least 70 years without prior dementia or cardiovascular disease. Fifteen baseline chronic conditions were grouped using latent class analysis (LCA). Observed-to-expected (O/E) ratios characterised conditions over-represented within clusters, and Cox proportional hazards models assessed associations with all-cause mortality. Results: Among 16,095 participants (mean age 74 years), 88.3% had multimorbidity at baseline; 4,217 deaths occurred over a median follow-up of 10.85 years. Five clusters were identified overall: hypertension and dyslipidemia (52.1%), gout and metabolic (14.4%), depressive symptoms, osteoporosis and frailty (10.0%), anaemia and kidney disease (10.2%), and hypotension, thyroid disorder and past cancer (13.3%). Sex-stratified analyses revealed three clusters in males and four in females. The frailty, depressive symptoms and osteoporosis cluster was associated with higher mortality in both sexes (aHR 1.56 [95% CI 1.40-1.73] in males; 1.68 [1.49-1.89] in females). Higher mortality was also observed for the metabolic, gout and kidney disease cluster in males (aHR 1.63 [1.47-1.81]) and the gout, anaemia and kidney disease cluster in females (aHR 1.96 [1.74-2.21]). Conclusions: Distinct multimorbidity clusters differed by sex and were associated with increased all-cause mortality. These findings may support risk stratification, targeted screening, and more person-centred management of older adults with multimorbidity.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2606.19808

Think Again or Think Longer? Selective Verification for Budget-Aware Reasoning

作者:

Sajib Acharjee Dip↗Dawei Zhou↗Liqing Zhang↗

Test-time reasoning is increasingly used as a serving-time control knob, but extra reasoning is not uniformly valuable: it can repair failed attempts, waste compute on already-correct answers, or introduce harmful answer changes. We study this as a deployment allocation problem rather than a new-verifier problem. We introduce \sevra, Selective Verification for Reasoning Allocation, a serving-layer controller that decides whether to preserve a frozen solver's initial answer or invoke active verification. Using a frozen Qwen3-4B solver, we log intervention outcomes and train recoverability-aware gates from serving-visible attempt state. On \mathfive, selective verification reaches 76.3\% accuracy, compared with 75.5\% for always verifying, while reducing post-generation tokens by 26.8\% and harmful flips from 2.2\% to 1.0\%. However, an 8,192-token initial solve reaches 76.0\% accuracy with 28\% fewer total model tokens, showing that selective recovery is useful but not the best tested cost frontier. In frozen transfer to \gsm, the selective policy verifies only 3.0\% of examples, improves accuracy from 93.4\% to 94.5\%, and reduces verification tokens by 91.2\% relative to always verifying; again, a longer initial solve matches its accuracy with fewer realized tokens. On CommonsenseQA, always-on verification hurts, while Self-Consistency@5 improves accuracy at about five times the realized token cost. The resulting deployment rule is: tune the initial budget first, then use selective recovery when explicit checks, bounded retries, auditability, or regression-risk control matter.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.16193

Cascaded Sparse Autoencoders Learn Multi-Level Visual Concepts in Multimodal LLMs

作者:

Yusong Zhao↗Hengyi Wang↗Tanuja Ganu↗Akshay Nambi↗Hao Wang↗

Multimodal Large Language Models (MLLMs) have demonstrated strong performance on vision-language tasks, yet their internal visual representations remain difficult to interpret. Sparse Autoencoders (SAEs) provide a scalable way to decompose dense model activations into sparse, interpretable features. However, existing SAE architectures primarily recover flat feature dictionaries and are less suited for explicit multi-level concept organization. In this paper, we introduce cascaded sparse autoencoders (CSAEs) for learning hierarchical visual concepts in MLLMs. Rather than nesting or stacking SAE sparse activation codes, CSAEs train a second-level SAE directly on the decoder weights of the first-level SAE, treating learned low-level feature directions as inputs for higher-level abstraction. This design enables CSAEs to learn "concepts of concepts" while avoiding drawbacks from the shared-prefix coupling of nesting, Matryoshka-style hierarchies and the bottlenecks of naively stacked SAEs. Experiments across Qwen3-VL, Gemma-3, and LLaVA on multiple visual datasets show that CSAEs improve interpretability in terms of hierarchical concept coherence over state-of-the-art SAE baselines. Results on concept steering further demonstrate that the learned concept groups support effective group-level interventions in MLLM outputs.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.16436

V2P-Manip: Learning Dexterous Manipulation from Monocular Human Videos

作者:

Kaihan Chen↗Yanming Shao↗Haifeng Ji↗Xiaokang Yang↗Yao Mu↗

Achieving autonomous robotic dexterous manipulation requires precise, human-like action sequences at scale. As a scalable supplement to costly teleoperation data, extracting trajectories with both visual fidelity and physical plausibility from monocular videos represents a promising frontier in embodied AI. To this end, we introduce V2P-Manip, an efficient framework designed to learn dexterous manipulation policies directly from human demonstration videos. We establish an efficient, integrated pipeline encompassing 3D asset acquisition, trajectory estimation, and dexterous policy learning. To bridge the gap between visual perception and physical constraints, we introduce a two-stage refinement process to enforce spatial alignment and physical consistency. Evaluations on the TACO and OakInk benchmarks demonstrate that our approach significantly outperforms previous methods in pose accuracy, adaptability to unstructured environments, and training efficiency. Ultimately, experimental results confirm an average success rate of over 75% across multiple synthetic manipulation tasks and validate the adaptability of the extracted manipulation priors across diverse dexterous hand embodiments.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16910

IMPACTeen: Intentions, Manipulation, Persuasion, Annotations, and Consequences in Teen Communication Dataset

作者:

Aleksander Szcz\k{e}sny↗Wiktoria Mieleszczenko-Kowszewicz↗Maciej Markiewicz↗Beata Bajcar↗Tomasz Adamczyk↗Jolanta Babiak↗Grzegorz Chodak↗Przemys{\l}aw Kazienko↗

IMPACTeen is a dataset of textual social influence scenarios spanning interpersonal, media-based, and digital settings in an adolescent context. It contains 1,021 texts, 5,100 individual annotation records, and gold labels for social influence techniques, with each text annotated from five distinct perspectives: teenagers, parents, psychologists, communication experts, and teachers. The resource was constructed through constrained LLM generation, followed by a two-step human editing and validation phase aimed at ensuring youth-context realism. A multi-dimensional annotation covered influence presence, techniques, intentions, consequences, resistance, reactions, and annotation confidence. The dataset supports research on social influence detection, annotator disagreement, cross-lingual modeling, and the training and evaluation of language models. The dataset was created in Polish and is accompanied by a corresponding English version.

阅读与讨论 → 访问原文 →

10.

arXiv (quant-ph) 2026-06-15 DOI: arXiv:2602.21689

Landscape-Similarity-Guided Optimization in Divide-and-Conquer QAOA

作者:

Sokea Sang↗Leanghok Hour↗Sanghyeon Lee↗Aniket Patra↗Hee Chul Park↗Moon Jip Park↗Youngsun Han↗

arXiv:2602.21689v3 Announce Type: replace Abstract: Divide-and-conquer strategies mitigate hardware constraints for the Quantum Approximate Optimization Algorithm (QAOA) on Noisy Intermediate-Scale Quantum (NISQ) devices by partitioning large interaction graphs into smaller, hardware-compatible sub-problems. However, this approach introduces a severe classical training bottleneck: a decomposition across $m$ boundary nodes generates $2^m$ distinct sub-problems that typically require independent optimization. In this work, we demonstrate that across diverse synthetic and real-world interaction graphs, the variational landscapes of these reduced QAOA instances actually exhibit a robust universality. Adapting the replica-overlap framework of spin-glass physics, we define a landscape-overlap order parameter $q$ to quantify geometric correlations between energy landscapes, revealing a sharp landscape-similarity transition as graph connectivity is tuned. Exploiting this, we introduce Doubly Optimized QAOA (DO-QAOA), an adaptive pipeline that collapses the sub-problems from $2^m$ distinct sub-problems into $K=\mathcal{O}(1)$ effective landscape classes. By performing optimization on a single representative sub-problem and dynamically transferring parameters to remaining sub-problems, DO-QAOA lowers runtime and quantum measurement overhead by orders of magnitude while maintaining a competitive Approximation Ratio Gap (ARG).

阅读与讨论 → 访问原文 →

11.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2606.12609

Viral Proteins Reveal Geometry of Protein Language Models

作者:

Arthur Bigot↗Harmon Bhasin↗Core Francisco Park↗Eugene Shakhnovich↗Dianzhuo Wang↗

arXiv:2606.12609v1 Announce Type: new Abstract: Protein language models are trained on highly imbalanced datasets, raising the question of how they represent underrepresented biological sequences. Using viral proteins as a case study across ESM model families, we identify a dominant nativeness axis in embedding space, aligned with masked reconstruction perplexity, that orders sequences from well-modeled cellular proteins through viral proteins to shuffled and random sequences. Scaling contracts this axis unevenly across viral families. Despite this, protein language model embeddings retain viral-specific signal: viral proteins remain linearly separable beyond zero-shot perplexity and shallow sequence features. Together, these results suggest that pLM representations are structured by a general notion of nativeness while preserving information specific to distinct biological groups.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2606.17000

The Complexity of Min-Max Optimization for Quadratic Polynomials

作者:

Martino Bernasconi↗Matteo Castiglioni↗Andrea Celli↗Alexandros Hollender↗

arXiv:2606.17000v1 Announce Type: cross Abstract: We prove that computing approximate stationary points of min-max optimization over the hypercube is PPAD-hard for quadratic polynomials. This holds even when the polynomials are multilinear, each variable appears in at most three monomials, and the approximation factor is inverse polynomial. As a direct consequence, we obtain the first PPAD-hardness results for two-team zero-sum polymatrix games.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.15527

Selective Synergistic Learning for Video Object-Centric Learning

作者:

WonJun Moon↗Jae-Pil Heo↗

Typical video object-centric learning (VOCL) approaches employ slot-based frameworks that rely on reconstruction-driven encoder-decoder architectures, where learning is mediated by two spatial maps: attention maps from the encoder and object maps from the decoder. As these two distinct maps exhibit different properties, a recent dense alignment strategy attempted to reconcile this discrepancy by enforcing agreement across all spatio-temporal patches via contrastive learning. However, this indiscriminate alignment inadvertently propagates the inherent weaknesses of each module, such as noisy encoder predictions and blurred decoder boundaries. Moreover, computing dense similarities across all pairs incurs a computational cost quadratic in the total number of spatio-temporal patches, severely limiting scalability. Motivated by this, we propose Selective Synergistic Learning (SSync). Instead of exhaustive patch-to-patch alignment, SSync prevents error propagation by selectively distilling only the most reliable cues: leveraging the encoder strictly for boundary refinement and the decoder for interior denoising. This is realized via a pseudo-labeling with linear complexity, eliminating the need for quadratic spatial comparisons. Also, to prevent the reinforcement of architectural biases like slot redundancy, we introduce a transitive pseudo-label merging that consolidates overlapping slots based on spatio-temporal activation consistency. Extensive studies demonstrate that SSync improves decomposition quality and serves as a versatile, plug-and-play module while also exhibiting exceptional robustness to slot configurations. Code is available at github.com/wjun0830/SSync.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2601.19099

m2sv: A Scalable Benchmark for Map-to-Street-View Spatial Reasoning

作者:

Yosub Shin↗Michael Buriek↗Igor Molybog↗

Vision–language models (VLMs) achieve strong performance on many multimodal benchmarks but remain brittle on spatial reasoning tasks that require aligning abstract overhead representations with egocentric views. We introduce m2sv, a scalable benchmark for map-to-street-view spatial reasoning that asks models to infer camera viewing direction by aligning a north-up overhead map with a Street View image captured at the same real-world intersection. We release m2sv-20k, a geographically diverse benchmark with controlled ambiguity, along with m2sv-sft-11k, a curated set of structured reasoning traces for supervised fine-tuning. Despite strong performance on existing multimodal benchmarks, the best evaluated VLM achieves only 65.2% accuracy on m2sv, below human annotators who reach 72.0% on average (and 95% for an expert) with strong inter-annotator agreement ($\kappa$ up to 0.76). While supervised fine-tuning and reinforcement learning yield consistent gains, cross-benchmark evaluations reveal limited transfer. Beyond aggregate accuracy, we systematically analyze difficulty in map-to-street-view reasoning using both structural signals and human effort, and conduct an extensive failure analysis of adapted open models. Our findings highlight persistent gaps in geometric alignment, evidence aggregation, and reasoning consistency, motivating future work on grounded spatial reasoning across viewpoints.

阅读与讨论 → 访问原文 →

15.

arXiv (math.PR) 2026-06-19 DOI: arXiv:2509.08629

A Cycle Walk for Sampling Measures on Spanning Forests for Redistricting

作者:

Daryl R. DeFord↗Gregory Herschlag↗Jonathan C. Mattingly↗

arXiv:2509.08629v2 Announce Type: replace-cross Abstract: We introduce the Cycle Walk, a new Markov chain Monte Carlo method for sampling distributions on balanced graph partitions, motivated by applications in political redistricting. The method operates on spanning forests and combines two types of updates: local "cycle" moves within districts and global moves that exchange population between adjacent districts while preserving balance constraints. This construction enables efficient Metropolis–Hastings correction while allowing proposals at multiple spatial scales. We show that the Cycle Walk naturally interpolates between existing approaches based on local updates and a class of global update methods derived from recombination (RECOM). Through a range of numerical experiments on synthetic graphs and real-world precinct data, we demonstrate that the Cycle Walk exhibits improved empirical convergence diagnostics for distributions that place weaker weight on spanning-tree counts, a regime that is challenging for existing methods. In particular, the algorithm remains effective when incorporating alternative compactness measures that more closely reflect policy-relevant criteria. These results suggest that the Cycle Walk provides a flexible and computationally efficient framework for sampling from a broader class of redistricting distributions than previously accessible with MCMC techniques.

阅读与讨论 → 访问原文 →

16.

arXiv (math.PR) 2026-06-12 DOI: arXiv:2606.12720

On McDiarmid's Inequality under Dependence via Approximate Tensorization of Entropy

作者:

Valentin Roth↗

arXiv:2606.12720v1 Announce Type: new Abstract: We argue that dependent versions of McDiarmid's inequality are a useful but underutilized tool in mathematical statistics, learning theory and theoretical computer science. To make this point, we first highlight that approximate tensorization of entropy (ATE) implies McDiarmid's via the Entropy Method. Second, we derive McDiarmid's inequality for non-isotropic Gaussian random vectors $X \sim \mathcal N(\mu, \Sigma)$ through ATE with a constant of the order of the condition number of $\Sigma$. We both independently obtain this ATE through a simple application of stochastic localization and also discuss how a more general ATE for the Gibbs sampler due to Ascolani et al., 2026 generalizes McDiarmid's-like concentration to strongly log-concave and log-smooth probability measures. We then apply the resulting concentration inequalities to resolve a question on the concentration of $\operatorname{sign}(X)$ posed by Simone Bombari, investigate Erdős-Rényi graphs under dependence and prove a Dvoretzky-Kiefer-Wolfowitz-type inequality for observations from a joint measure fulfilling ATE and continuous marginal CDFs. For the class of strongly log-concave and log-smooth measures, this result improves upon a prior Dvoretzky-Kiefer-Wolfowitz-type inequality for non-i.i.d. observations due to Bobkov and Götze, 2010, by establishing the expected $1/\sqrt{n}$-rate of convergence under weak dependence instead of $n^{-1/3}$.

阅读与讨论 → 访问原文 →

17.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2602.00845

Optimizing Agentic Reasoning with Retrieval via Synthetic Semantic Information Gain Reward

作者:

Senkang Hu↗Yong Dai↗Yuzhi Zhao↗Yihang Tao↗Yu Guo↗Zhengru Fang↗Sam Tak Wu Kwong↗Yuguang Fang↗

arXiv:2602.00845v3 Announce Type: replace Abstract: Agentic reasoning enables large reasoning models (LRMs) to dynamically acquire external knowledge, but yet optimizing the retrieval process remains challenging due to the lack of dense, principled reward signals. In this paper, we introduce InfoReasoner, a unified framework that incentivizes effective information seeking via a synthetic semantic information gain reward. Theoretically, we redefine information gain as uncertainty reduction over the model's belief states, establishing guarantees, including non-negativity, telescoping additivity, and channel monotonicity. Practically, to enable scalable optimization without manual retrieval annotations, we propose an output-aware intrinsic estimator that computes information gain directly from the model's output distributions using semantic clustering via bidirectional textual entailment. This intrinsic reward guides the policy to maximize epistemic progress, enabling efficient training via Group Relative Policy Optimization (GRPO). Experiments across seven question-answering benchmarks demonstrate that InfoReasoner consistently outperforms strong retrieval-augmented baselines, achieving up to 5.4% average accuracy improvement. Our work provides a theoretically grounded and scalable path toward agentic reasoning with retrieval. The code is available at https://github.com/dl-m9/InfoReasoner

阅读与讨论 → 访问原文 →

18.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.19597

PrefSQA: Pairwise Preference Prediction for Speech Quality Assessment and the Critical Role of High Quality Datasets

作者:

Junyi Fan↗Donald S. Williamson↗

arXiv:2606.19597v1 Announce Type: cross Abstract: Mean opinion scores (MOS) are widely used for speech quality assessment, yet scalar labels are sensitive to rater variability and listening test differences. This introduces labeling noise, which limits the reliability of MOS prediction. Preference prediction reduces this variability as listeners compare signals directly, producing cleaner labels. We study MOS-free preference prediction and propose PrefSQA, which incorporates uncertainty-aware logits, an impairment attention head, and a module based on non-matching-reference comparisons. We use and refine five datasets, including MOS-derived and low-noise simulated sets with matching and non-matching content, experiment with human preference sets, and test on unseen data. Experiments show small improvements on MOS-derived data, while other sets reveal clear improvement over the baselines, highlighting the value of high-quality preference data and demonstrating the effectiveness of the proposed method.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2606.18613

Are LLMs Ready to Assist Physicians? PhysAssistBench for Interactive Doctor-Patient-EHR Assistance

作者:

Tianming Du↗Peijie Yu↗Sihan Shang↗Danli Shi↗My Linh Nguyen↗Shengbo Gao↗Guangyuan Li↗Yinghong Yu↗Yan Jiang↗Qianlong Zhao↗Behzad Bozorgtabar↗Shaoxiong Ji↗…

The most plausible near-term role of medical LLMs is to assist rather than replace physicians, yet current evaluations often test isolated capabilities: clinical knowledge, EHR system interaction, or patient communication. Physician assistance instead requires coordinating these capabilities within the same interaction, where physicians issue underspecified requests, patients describe symptoms ambiguously, and EHR systems demand precise tool use. We introduce PhysAssistBench, a benchmark for interactive doctor-patient-EHR assistance. Built from real MIMIC-IV cases, PhysAssistBench uses a scalable pipeline to construct agentic patients: interactive, record-grounded agents that turn static EHR records into multi-turn clinical scenarios while preserving clinical factuality. PhysAssistBench provides a curated bilingual evaluation set of 1,296 manually reviewed and physician-validated turns. Experiments with leading LLMs show that current models remain unreliable in this setting, which exposes a key bottleneck for clinical LLMs: reliable assistance requires coordination across knowledge, communication, and systems, not isolated gains in any of them.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.19489

Concept Flow Models: Anchoring Concept-Based Reasoning with Hierarchical Bottlenecks

作者:

Ya Wang↗Adrian Paschke↗

arXiv:2606.19489v1 Announce Type: cross Abstract: Concept Bottleneck Models (CBMs) enhance interpretability by projecting learned features into a human-understandable concept space. Recent approaches leverage vision-language models to generate concept embeddings, reducing the need for manual concept annotations. However, these models suffer from a critical limitation: as the number of concepts approaches the embedding dimension, information leakage increases, enabling the model to exploit spurious or semantically irrelevant correlations and undermining interpretability. In this work, we propose Concept Flow Models (CFMs), which replace the flat bottleneck with a hierarchical, concept-driven decision tree. Each internal node in the hierarchy focuses on a localized subset of discriminative concepts, progressively narrowing the prediction scope. Our framework constructs decision hierarchies from visual embeddings, distributes semantic concepts at each hierarchy level, and trains differentiable concept weights through probabilistic tree traversal. Extensive experiments on diverse benchmarks demonstrate that CFMs match the predictive performance of flat CBMs, while substantially mitigating information leakage by reducing effective concept usage. Furthermore, CFMs yield stepwise decision flows that enable transparent and auditable model reasoning with hierarchical class structures.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.CV) 2026-06-11 DOI: arXiv:2603.00461

ReMoT: Reinforcement Learning with Motion Contrast Triplets

作者:

Cong Wan↗Zeyu Guo↗Jiangyang Li↗SongLin Dong↗Yifan Bai↗Lin Peng↗Zhiheng Ma↗Yihong Gong↗

We present ReMoT, a unified training paradigm to systematically address the fundamental shortcomings of VLMs in spatio-temporal consistency – a critical failure point in navigation, robotics, and autonomous driving. ReMoT integrates two core components: (1) A rule-based automatic framework that generates ReMoT-16K, a large-scale (16.5K triplets) motion-contrast dataset derived from video meta-annotations, surpassing costly manual or model-based generation. (2) Group Relative Policy Optimization, which we empirically validate yields optimal performance and data efficiency for learning this contrastive reasoning, far exceeding standard Supervised Fine-Tuning. We also construct the first benchmark for fine-grained motion contrast triplets to measure a VLM's discrimination of subtle motion attributes (e.g., opposing directions). The resulting model achieves state-of-the-art performance on our new benchmark and multiple standard VLM benchmarks, culminating in a remarkable 25.1% performance leap on spatio-temporal reasoning tasks.

阅读与讨论 → 访问原文 →

22.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.18413

Searching for Synergy in Shared Workspace Human-AI Collaboration

作者:

Nachiket Kotalwar↗Rohini Das↗Carolyn Rose↗

arXiv:2606.18413v1 Announce Type: new Abstract: Automated AI agents are increasingly capable, yet many scientific and professional tasks require human judgment and contextual expertise. We study shared-workspace human-AI teams, where AI agents and human collaborators must coordinate responsibilities before submitting a final answer. Using the Collaborative Gym environment with DiscoveryBench tasks, we examine when adding simulated human collaborators improves performance and when process loss turns additional collaborators into coordination overhead. Across 1,482 sessions, adding relevant collaborators can lower performance when teams lack structure to coordinate their contributions. We then evaluate scaffolding that combines shared group memory with simulated human-in-the-loop (HITL) gates, where selected actions require approval from a designated simulated participant. This scaffolding yields higher mean performance, most clearly in three-person teams, with clearer responsibility signals and stronger routing of expertise to team actions. Overall, how human-AI teams coordinate and integrate expertise matters as much as the capability available to them.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2603.14407

Towards One-for-All Anomaly Detection for Tabular Data

作者:

Shiyuan Li↗Yixin Liu↗Yu Zheng↗Xiaofeng Cao↗Shirui Pan↗Heng Tao Shen↗

arXiv:2603.14407v2 Announce Type: replace Abstract: Tabular anomaly detection (TAD) aims to identify samples that deviate from the majority in tabular data and is critical in many real-world applications. However, existing methods follow a ``one model for one dataset (OFO)'' paradigm, which relies on dataset-specific training and thus incurs high computational cost and yields limited generalization to unseen domains. To address these limitations, we propose OFA-TAD, a generalist one-for-all (OFA) TAD framework that only requires one-time training on multiple source datasets and can generalize to unseen datasets from diverse domains on-the-fly. To realize one-for-all tabular anomaly detection, OFA-TAD extracts neighbor-distance patterns as transferable cues, and introduces multi-view neighbor-distance representations from multiple transformation-induced metric spaces to mitigate the transformation sensitivity of distance profiles. To adaptively combine multi-view distance evidence, a Mixture-of-Experts (MoE) scoring network is employed for view-specific anomaly scoring and entropy-regularized gated fusion, with a multi-strategy anomaly synthesis mechanism to support training under the one-class constraint. Extensive experiments on 34 datasets from 14 domains demonstrate that OFA-TAD achieves superior anomaly detection performance and strong cross-domain generalizability under the strict OFA setting. The source code is available at https://github.com/Shiy-Li/OFA-TAD.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2505.09655

DRA-GRPO: Your GRPO Needs to Know Diverse Reasoning Paths for Mathematical Reasoning

作者:

Xiwen Chen↗Wenhui Zhu↗Peijie Qiu↗Xuanzhao Dong↗Hao Wang↗Haiyu Wu↗Huayu Li↗Aristeidis Sotiras↗Yalin Wang↗Abolfazl Razi↗

Post-training LLMs with Reinforcement Learning, specifically Group Relative Policy Optimization (GRPO), has emerged as a paradigm for enhancing mathematical reasoning. However, standard GRPO relies on scalar correctness rewards that are often non-injective with respect to semantic content: distinct reasoning paths receive identical rewards. This leads to a Diversity-Quality Inconsistency, where the policy collapses into a narrow set of dominant modes while ignoring equally valid but structurally novel strategies. To bridge this gap, we propose Diversity-aware Reward Adjustment (DRA), a theoretically grounded framework that calibrates the reward signal using the semantic density of sampled groups. By leveraging Submodular Mutual Information (SMI), DRA implements an Inverse Propensity Scoring (IPS) mechanism that effectively de-biases the gradient estimation. This creates a repulsive force against redundancy, driving the policy to achieve better coverage of the high-reward landscape. Our method is plug-and-play and integrates seamlessly with GRPO variants. Empirical evaluations on five math benchmarks demonstrate that DRA-GRPO consistently outperforms strong baselines, achieving an average accuracy of 58.2% on DeepSeek-R1-Distill-Qwen-1.5B with only 7,000 training samples and $55 cost, highlighting the critical role of diversity calibration in data-efficient alignment. The code is available at https://github.com/xiwenc1/DRA-GRPO.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.17053

Context-Aware RL for Agentic and Multimodal LLMs

作者:

Peiyang Xu↗Bangzheng Li↗Sijia Liu↗Karthik R. Narasimhan↗Pramod Viswanath↗Prateek Mittal↗Xingyu Fu↗

Large language models (LLMs) often fail when answering requires identifying a small but decisive piece of evidence within a long or complex context, such as a single line in a tool trace or a subtle detail in an image. We propose ContextRL, a context-aware reinforcement learning (RL) method that improves long-horizon reasoning and multimodal performance through an indirect auxiliary objective. Instead of supervising only the final answer, ContextRL presents the model with a query, an answer, and two highly similar contexts, and rewards it for selecting the context that supports the query–answer pair, thereby encouraging fine-grained grounding. We construct contrastive context data in two domains: for coding agents, trajectories serve as contexts, yielding 1k pairs built via condition filtering; for multimodal reasoning, images serve as contexts, yielding 7K pairs built via generative editing and similarity search. ContextRL achieves average gains of +2.2% over standard GRPO on 5 long-horizon benchmarks, and +1.8% across 12 diverse visual question answering benchmarks. To disentangle the effect of the proposed objective from that of additional data, we compare against data-augmentation baselines that repurpose the same contrastive contexts as standard query–context–answer examples. These baselines provide little to no improvement, showing that the gains arise from the proposed context-selection objective rather than from the contrastive data alone.

阅读与讨论 → 访问原文 →