中 EN
登录 注册账户

Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达，利用大语言模型自动生成交叉领域文献分析简报。

登录 注册账户
01.
arXiv (CS.CL) 2026-06-17

Smarter edits? Post-editing with error highlights and translation suggestions

作者:
Fleur V. J. van Tellingen↗Gautam Ranka↗Dora \v{Z}ug\v{c}i\'c↗Joyce van der Wal↗Andrea Camasta↗Livio Guerra↗Alina Karakanta↗

As MT quality increases, interest in enhanced post-editing features such as QE-derived error highlights is growing, yet evidence for their usefulness remains limited. In this work, we explore the usefulness of LLM-derived error highlights and correction suggestions based on automatic post-editing (APE). We conduct a study where professional translators (En-Nl) post-edit translations using APE error highlights and correction suggestions and compare productivity, quality and user experience to regular PE and PE with QE-derived highlights. While no condition yielded productivity or quality gains compared to regular PE, APE highlights were better received than QE-derived highlights, and correction suggestions improved overall user experience.

阅读与讨论 → 访问原文 →
02.
arXiv (CS.CL) 2026-06-16

Generative causal testing to bridge data-driven models and scientific theories in language neuroscience

作者:
Richard Antonello↗Chandan Singh↗Shailee Jain↗Aliyah Hsu↗Sihang Guo↗Jianfeng Gao↗Bin Yu↗Alexander Huth↗

Representations from large language models are highly effective at predicting BOLD fMRI responses to language stimuli. However, these representations are largely opaque: it is unclear what features of the language stimulus drive the response in each brain area. We present generative causal testing (GCT), a framework for generating concise explanations of language selectivity in the brain from predictive models and then testing those explanations in follow-up experiments using LLM-generated stimuli.This approach is successful at explaining selectivity both in individual voxels and cortical regions of interest (ROIs), including newly identified microROIs in prefrontal cortex. We show that explanatory accuracy is closely related to the predictive power and stability of the underlying predictive models. Finally, we show that GCT can dissect fine-grained differences between brain areas with similar functional selectivity. These results demonstrate that LLMs can be used to bridge the widening gap between data-driven models and formal scientific theories.

阅读与讨论 → 访问原文 →
03.
arXiv (quant-ph) 2026-06-12

Proper and improper mixed states serve as different prior beliefs for quantum state retrodiction

作者:
Mingxuan Liu↗Valerio Scarani↗Ge Bai↗

arXiv:2502.10030v2 Announce Type: replace Abstract: A mixed quantum state can be taken as capturing an unspecified form of ignorance; or as describing the lack of knowledge about the true pure state of the system ("proper mixture"); or as arising from entanglement with another system that has been disregarded ("improper mixture"). These different views yield identical density matrices and therefore identical predictions for future measurements. But when used as prior beliefs for inferring the past state from later observations ("retrodiction"), they lead to different updated beliefs. This is a purely quantum feature of Bayesian agency. Based on this observation, we establish a framework for retrodicting on any quantum belief and we prove a necessary and sufficient condition for the equivalence of beliefs. We also illustrate how these differences have operational consequences in quantum state recovery.

阅读与讨论 → 访问原文 →
04.
arXiv (CS.CL) 2026-06-12

G-Long: Graph-Enhanced Memory Management for Efficient Long-Term Dialogue Agents

作者:
Minjun Choi↗Yoonjin Jang↗Sangwon Youn↗Youngjoong Ko↗

While Large Language Models (LLMs) have advanced open-domain dialogue systems, maintaining long-term consistency remains a challenge due to inherent limitations in long-context reasoning and the inefficiency of processing extensive raw text. Existing approaches typically rely on either unstructured memory storage, which is prone to information loss, or computationally expensive LLMs that incur high latency. To address these limitations, we propose G-Long, a graph-enhanced framework that utilizes a fine-tuned small Language Model (sLM) for structured triplet extraction and associative retrieval, significantly reducing operational costs. Furthermore, we introduce the novel attention-aware importance scoring mechanism that leverages the intrinsic cross-attention signals of a T5 summarizer to identify salient memories. Extensive experiments across diverse benchmarks demonstrate that G-Long achieves state-of-the-art performance in both response generation and memory retrieval, yielding performance gains of up to 9.8% in response quality on MSC and 40.8% in retrieval recall on LME, while significantly minimizing computational overhead.

阅读与讨论 → 访问原文 →
05.
arXiv (CS.CL) 2026-06-11

WorldReasoner: Evaluating Whether Language Model Agents Forecast Events with Valid Reasoning

作者:
Yizhou Chi↗Eric Chamoun↗Zifeng Ding↗Andreas Vlachos↗

Forecasting real-world events requires language-model agents to reason under uncertainty from incomplete, time-bounded information. Yet evaluating whether agents genuinely forecast requires more than final-answer accuracy: a model may be correct by recalling memorized training facts, citing fabricated evidence, or producing an unsupported causal story. We present WorldReasoner, an evaluation framework for temporally valid event forecasting. Each task gives an agent a resolved forecasting question, a simulated forecast date, and access only to evidence available before that date; after resolution, the framework scores the submitted probability, cited evidence, and optional causal event graph. WorldReasoner reports three complementary axes: outcome quality against resolved answers, evidence quality over cited sources, and reasoning quality against post-resolution hindsight graphs. The benchmark is built by an agentic construction pipeline that generates forecasting questions, collects time-stamped evidence, and builds hindsight reference graphs at scale, yielding 345 resolved tasks derived from 14,141 articles with graphs covering 8,087 extracted events. Across six controlled agent settings, temporally valid retrieval is the strongest driver of outcome accuracy; causal graph construction improves key-event recovery; and correct graph-enabled forecasts are more strongly grounded in key events and relevant sources, yet agents still struggle to convert grounded evidence into calibrated probabilities.

阅读与讨论 → 访问原文 →
06.
arXiv (CS.CL) 2026-06-19

Pitch Spelling Jazz Lead Sheets, Solo Transcriptions, Classical Piano and Monophonic Scores

作者:
Augustin Bouquillard↗Florent Jacquemard↗

We present an algorithm for pitch spelling and key estimation. Given an input in MIDI-like format, containing information on note pitches (expressed in semitones relative to the lowest reference note) and bar boundaries, it estimates the appropriate note names, a global Key Signature, and a local scale for each bar. This related information elements are evaluated jointly during two stages of optimisation. During an initial 'modal' stage, a probable scale is proposed for each bar, minimising the number of accidentals to be printed in the printed score with a shortest-path search. Then, during a second stage called 'tonal', these local scales are used to estimate the Key Signature and note names that would result in the best musical notation for the entire piece. We present evaluations conducted on datasets comprising a variety of digital musical scores: jazz lead sheets taken from the Real Book, transcriptions of recordings of jazz soli and bass lines, traditional tunes, as well as classical scores for piano and monophonic instruments. Our procedure was originally designed for use in music transcription, specifically for building digital collections of jazz solos transcribed from audio recordings, for the purposes of music analysis, teaching and the preservation of cultural heritage. This method should also prove useful for other tasks related to the processing of musical notation. Furthermore, to this end, we have defined new distances between various common jazz scales, which may be of some interest to musicological studies.

阅读与讨论 → 访问原文 →
07.
arXiv (CS.CL) 2026-06-18

EARS: Explanatory Abstention for Reliable Sub-Agent Modeling in Large-scale Multi-Agent Systems

作者:
Shuang Xie↗Yunan Lu↗Han Li↗Lingyun Wang↗

In large-scale enterprise settings, centralized multi-agent systems (MAS) are increasingly adopted, in which a coordinator delegates user requests to lightweight, domain-specialized sub-agents. While this architecture improves modularity, scalability, and cost efficiency, its reliability depends not only on accurate routing but also on sub-agents' ability to calibrate their responses to capability constraints. In particular, sub-agents built on smaller fine-tuned models often struggle with such calibration, leading them to over-answer ambiguous, underspecified, misrouted, or unsupported requests and produce hallucinated outputs instead of actionable feedback. To address this challenge, we present EARS (Explanatory Abstention for Reliable Sub-Agent Modeling), a production-oriented framework that reframes sub-agent abstention as an inter-agent communication protocol: a sub-agent does not merely abstain, but exposes an actionable failure state to the coordinator. EARS curates human-agent interaction data using an ensemble of calibrated LLM-as-a-Judge models, producing structured abstention labels and rationales under a taxonomy of sub-agent failure modes. These data are used to fine-tune sub-agents to detect failure conditions and return rationales for coordinator-level clarification, rerouting, or fallback. We evaluate EARS in a large-scale production e-commerce assistant supporting enterprise business intelligence workflows. EARS improves the overall response pass rate from 68.5% to 78.9%, demonstrating that sub-agent-side explanatory abstention improves MAS reliability.

阅读与讨论 → 访问原文 →
08.
PLOS Computational Biology 2026-06-01

Histology-informed spatial domain identification through multi-view graph convolutional networks

作者:
Huihui Zhang↗

by Huihui Zhang, Jiaxing Chang, Zirong Li, Yue Sun, Pinli Hu, Haoxiu Wang, Hang Yang, Yonglin Ren, Xingtan Zhang, Zehua Chen, Kok Wai Wong, Haojing Shao Identifying spatial domains is crucial in spatial transcriptomics, yet effectively integrating gene expression, spatial location, and histology remains challenging. We present STESH, a Spatial Transcriptomics clustering method that combines Expression, Spatial information and Histology. STESH extracts histological features using a convolutional neural network and generates expression, histology, spatial, and collaborative convolution modules for a multi-view graph convolutional network with a decoder and attention mechanism. We evaluated STESH on multiple tissue types and technology platforms. STESH consistently outperformed ten state-of-the-art methods, achieving superior clustering accuracy with the highest scores in adjusted Rand index, normalized mutual information, and Fowlkes-Mallows index.

阅读与讨论 → 访问原文 →
09.
arXiv (CS.CL) 2026-06-12

Agent-based models for the evolution of morphological alternation patterns

作者:
Aravinth Kulanthaivelu↗Richard Sproat↗

Why is the past of English "go" the apparently unrelated "went"? Such alternations are frequent in languages. They neither aid communication nor learnability, yet they can be persistent, surviving over centuries or millennia. We present a multi-agent simulation of the emergence of morphological stem and inflection alternations. Alternate forms arise by phonological changes or, as with "go/went", from lexical alternatives associated with a subset of the population. When an agent 'hears' another agent use a novel form for a slot in the paradigm of a word (say, the past tense of go), they will with some probability adopt that form, possibly spreading its use to other slots in the paradigm that shared the same original form. Thus alternative forms can spread through the population and become entrenched as stem or inflectional marker alternants. Unlike many previous computational studies, our system allows for naturalistic lexical forms, realistic phonological rules, lexicons with hundreds or thousands of entries, and agent populations in the tens or hundreds. It supports several network topologies, diffusion patterns and agent adoption policies. One issue with such simulations is evaluation: how realistic is the resulting morphology compared to those of real languages? We introduce the AI Historical Linguist, a novel Large Language Model-driven system that models a debate between two historical linguists. We use this to compare a set of real language morphologies, disguised morphologies, and experimentally evolved morphologies. The results suggest that among the factors that favor more plausible morphologies are scale-free social networks and random Bernoulli adoption of forms. We also present three case studies modeling attested historical changes, allowing us to test what might have happened if history had been different. All code and data are released.

阅读与讨论 → 访问原文 →
10.
arXiv (CS.LG) 2026-06-17

Randomized Midpoint Method for Log-Concave Sampling under Constraints

作者:
Yifeng Yu↗Shijie Zhang↗Lu Yu↗

arXiv:2405.15379v3 Announce Type: replace-cross Abstract: In this paper, we study the problem of sampling from log-concave distributions supported on convex and compact sets, with a particular focus on the randomized midpoint discretization of both overdamped and kinetic Langevin diffusions in constrained domains. We revisit the proximal framework for handling constraints through projection operators and develop a more general formulation that encompasses Euclidean, Bregman, and Gauge projections. The resulting smooth approximation allows a unified and tractable analysis of Langevin algorithms and their variants under constraints. Within this framework, we establish convergence guarantees in Wasserstein-$q$ $(q\geqslant 1)$ distances between the smooth surrogate and the target distribution. We further derive complementary lower bounds, showing that the results are near-optimal in order. Building upon this tight approximation analysis, we obtain new convergence guarantees for the randomized midpoint Langevin algorithms and refined bounds for both vanilla and kinetic Langevin Monte Carlo methods under constraints, thereby advancing the theoretical understanding of constrained diffusion-based sampling.

阅读与讨论 → 访问原文 →
11.
arXiv (CS.AI) 2026-06-16

Heteroskedastic Signals in Budgeted LLM Verification: Structural Heterogeneity Limits Optimization Gains

作者:
Jinlong Yang↗

arXiv:2606.15841v1 Announce Type: new Abstract: Large language model (LLM) systems increasingly use uncertainty signals to allocate limited computation across verification, test-time scaling, tool execution, and other selective-compute decisions. Such policies rely on a global signal comparability assumption: equal scores should carry comparable decision value across inputs. Using budgeted verification as a controlled diagnostic setting, we identify a failure mode of this assumption: uncertainty quality is heteroskedastic across cost strata, with some regions exhibiting near-random discriminability despite concentrating many errors. Under an explicit local model, we characterize the resulting distortion of global allocation and show that its upper bound scales with cross-stratum signal-quality dispersion. We separate weak signals, optimization instability, and structural heterogeneity through a controlled intervention hierarchy: Threshold, MP-Adapt, MP-Strat, and a deliberately simple cost-stratified thresholding intervention (CST). Across MBPP and MATH using Qwen3-8B, LLaMA3-8B, and GPT-4o-mini, global online adaptation yields inconsistent gains over static thresholding; MP-Strat partially recovers performance, while CST improves hit rate by up to 17 percentage points in strongly heterogeneous settings without gradient updates. These results identify structural heterogeneity, rather than optimizer weakness alone, as the primary bottleneck in the observed settings. More broadly, misaligned feedback structure cannot always be repaired by stronger optimization.

阅读与讨论 → 访问原文 →
12.
arXiv (CS.AI) 2026-06-16

AutoDojo: Adaptive Attacks Expose Superficial Defenses and User-Underspecification Limits in LLM Agents

作者:
Xinhang Ma↗Taoran Li↗Chaowei Xiao↗Zhiyuan Yu↗Ning Zhang↗Yevgeniy Vorobeychik↗

arXiv:2606.15057v1 Announce Type: cross Abstract: Indirect prompt injection (IPI) is a major security threat to LLM-powered agents. Thus, a growing body of work have proposed a variety of defensive approaches against IPI. These can be grouped into three broad categories: 1) prompt-based (using prompting as a way to prevent agents from following malicious instructions), 2) detection-based (identifying and filtering malicious instructions), and 3) system-level (using systems insights, such as control and data isolation, for defense). However, commonly used benchmarks for evaluating defense, such as AgentDojo, are inherently static, generating a fixed distribution of IPI attacks. Consequently, static benchmarks do not usefully evaluate defense robustness to adaptive threats. We address this issue by developing AutoDojo, an adaptive extension of AgentDojo that optimizes IPI against a given defense. Using AutoDojo against state-of-the-art IPI defenses across three task suites and five target models, we make two key observations. First, many defenses offer only limited protection: a cheap, black-box adaptive attack using a frontier LLM to iteratively optimize the injection raises attack success rate (ASR) well above the level achieved by static injections against nearly all evaluated defenses. Against a filter that reduces static ASR to 0\%, AutoDojo recovers 28\% overall and 64\% on action-open tasks. Second, for prompt-level and filter-based defenses, ASR is substantially higher on action-open tasks – where the user's request delegates the action itself to attacker-controlled content – than on precisely specified tasks. This is a structural limit: on such tasks the injection can pose as ordinary data rather than an explicit instruction, bypassing defenses that rely on detecting instruction-like text. AutoDojo is publicly available at https://github.com/xhOwenMa/AutoDojo.

阅读与讨论 → 访问原文 →
13.
arXiv (CS.AI) 2026-06-18

Large-Scale OD Matrix Estimation with A Deep Learning Method

作者:
Zheli Xiong↗Defu Lian↗Enhong Chen↗Gang Chen↗Xiaomin Cheng↗

arXiv:2310.05753v2 Announce Type: replace Abstract: The estimation of origin-destination (OD) matrices is a crucial aspect of Intelligent Transport Systems (ITS). It involves adjusting an initial OD matrix by regressing the current observations like traffic counts of road sections (e.g., using least squares). However, the OD estimation problem lacks sufficient constraints and is mathematically underdetermined. To alleviate this problem, some researchers incorporate a prior OD matrix as a target in the regression to provide more structural constraints. However, this approach is highly dependent on the existing prior matrix, which may be outdated. Others add structural constraints through sensor data, such as vehicle trajectory and speed, which can reflect more current structural constraints in real-time. Our proposed method integrates deep learning and numerical optimization algorithms to infer matrix structure and guide numerical optimization. This approach combines the advantages of both deep learning and numerical optimization algorithms. The neural network(NN) learns to infer structural constraints from probe traffic flows, eliminating dependence on prior information and providing real-time performance. Additionally, due to the generalization capability of NN, this method is economical in engineering. We conducted tests to demonstrate the good generalization performance of our method on a large-scale synthetic dataset. Subsequently, we verified the stability of our method on real traffic data. Our experiments provided confirmation of the benefits of combining NN and numerical optimization.

阅读与讨论 → 访问原文 →
14.
bioRxiv (Bioinfo) 2026-06-20

Evaluation of Trypanosoma brucei Phosphofructokinase Allosteric Inhibition: An In-Silico Study

作者:
Gumbis↗Houston↗D. R↗

Human African trypanosomiasis, caused by a protozoan parasite Trypanosoma brucei, is a neglected tropical disease for which well-tolerated, conveniently administered, and highly efficacious medicines are still missing. Previously, T. brucei Phosphofructokinase was targeted by small-molecule inhibitor development efforts. This approach has shown promise both in vitro and in vivo. In this study, we have used these wet-lab results, evaluated the compounds already characterised by Molecular Dynamics simulations, found relationships between in silico and wet-lab data and used these observations to evaluate compounds that we selected through several different approaches of virtual screens. We observed that inhibitor-ATP interactions are highly predictive of the inhibitory activity. Several compounds selected through virtual screens have outperformed previously characterised compounds.

阅读与讨论 → 访问原文 →
16.
arXiv (CS.CL) 2026-06-16

BALTO: Balanced Token-Level Policy Optimization for Hallucination Mitigation

作者:
Ning Li↗Zixuan Guo↗Yan Xu↗Wenbo Fei↗Yifan Niu↗Chang Luo↗Yasheng Wang↗Weiwen Liu↗Yong Yu↗Weinan Zhang↗

Hallucinations remain a major obstacle to deploying large language models (LLMs) in knowledge-intensive settings, where generated responses must be faithfully grounded in provided evidence. Reinforcement learning (RL) is a promising direction for hallucination mitigation, but response-level faithfulness rewards suffer from a granularity mismatch: localized hallucinations can cause supported content to receive spurious penalties. Although recent work introduces fine-grained feedback such as claim-level verification and token-level rewards, unbalanced credit assignment can still induce length, verbosity, or optimization-noise biases. We propose BALTO, a Balanced Token-level Policy Optimization framework for hallucination mitigation. BALTO extracts checkable factual claims, verifies them against the reference context, and projects claim-level judgments to token-level labels. A balanced token-level credit assignment mechanism is introduced into the framework. This design redistributes probability mass from unsupported content toward faithful content, rather than suppressing the entire response. We systematically analyze the limitations of response-level rewards from a theoretical standpoint, and prove BALTO's advantages in training stability and optimization efficiency for hallucination mitigation. Experiments on ConFiQA, RAGTruth, and FinLLM-Eval show that BALTO achieves the highest faithfulness across all six model–benchmark settings and consistently outperforms existing post-training baselines in Q-Score, demonstrating a stronger faithfulness–informativeness trade-off.

阅读与讨论 → 访问原文 →
17.
arXiv (CS.AI) 2026-06-16

Beyond Classification: A Cough Regression Benchmark for Respiratory Acoustic Foundation Models

作者:
Mayur Sanap↗Prasanna Desikan↗Edgar Lobaton↗

arXiv:2606.15436v1 Announce Type: cross Abstract: Respiratory acoustic foundation models (FMs) excel at cough classification, yet their ability to predict continuous health quantities from cough audio remains largely unexplored, despite the clinical value of passive age, BMI, and disease probability estimation in settings where physical measurements are unavailable. We introduce the multi-model, multi-target cough regression benchmark evaluating five FMs (OPERA-CT, OPERA-CE, OPERA-GT, HeAR, M2D+Resp) across six targets on three datasets under subject-disjoint protocols, comparing linear, MLP-small, and full MLP regression heads. MLP-small beats the mean-predictor baseline on all tasks and linear probing in 23 of 30 model x task cases, with full MLP overfitting on small clinical data but recovering on larger sets, revealing a dataset size x head-capacity trade-off. HeAR leads within-dataset age regression on Coswara (9.12 yr MAE); its CIDRZ result is excluded from headline claims owing to possible HeAR-CIDRZ pretraining overlap. OPERA-GT is favored over OPERA-CT on age in all three datasets, with the CIDRZ margin within seed variance, extending a generative-pretraining advantage from breath to cough. HeAR and M2D+Resp reach near-full performance at N = 50 samples while OPERA models require N = 400. Cross-dataset transfer is strongly asymmetric as large diverse data generalises to small clinical populations (CoughVID to CIDRZ: -0.17 yr) but not vice versa (CIDRZ to Coswara: +2.43 yr, +26.6%).

阅读与讨论 → 访问原文 →
18.
arXiv (CS.AI) 2026-06-18

Skill-MAS: Evolving Meta-Skill for Automatic Multi-Agent Systems

作者:
Hehai Lin↗Qi Yang↗Chengwei Qin↗

arXiv:2606.18837v1 Announce Type: cross Abstract: Large Language Model (LLM)-based automatic Multi-Agent Systems (MAS) generation has become a crucial frontier for tackling complex tasks. However, existing methods face a dilemma between model capability and experience retention. Inference-time MAS leverages frozen frontier LLMs but repeats identical searches without learning from past experience. Conversely, Training-time MAS internalizes experience via gradient updates but is constrained by the low capability ceiling of smaller models, and is hard to scale to large frontier LLMs. To bridge this gap, we propose Skill-MAS, a novel third path that decouples experience retention from parametric updates by conceptualizing the high-level orchestration capability as an evolvable Meta-Skill. Skill-MAS refines this architectural knowledge through a closed optimization loop: (1) Multi-Trajectory Rollout samples a behavioral distribution for each task under the current Meta-Skill; and (2) Selective Reflection adaptively selects priority tasks and applies hierarchical contrastive analysis to distill systemic experience into generalizable, strategy-level principles. Extensive experiments across four complex benchmarks and four distinct LLMs demonstrate that Skill-MAS not only achieves remarkable performance gains but also maintains a favorable cost-performance trade-off. Further analysis reveals that the evolved Meta-Skills are highly robust and exhibit strong transferability across unseen tasks and different LLMs.

阅读与讨论 → 访问原文 →
19.
arXiv (CS.AI) 2026-06-18

Spotlight: Synergizing Seed Exploration and Spot GPUs for DiT RL Post-Training

作者:
Ruiqi Lai↗Dakai An↗Wei Gao↗Ju Huang↗Siran Yang↗Jiamang Wang↗Lin Qu↗Dmitrii Ustiugov↗Wei Wang↗

arXiv:2606.19004v1 Announce Type: cross Abstract: Reinforcement learning (RL) post-training of Diffusion Transformers (DiTs) is prohibitively expensive, requiring thousands of high-end GPUs. Existing works explore two directions to reduce cost: seed exploration improves training convergence by selecting high-contrast samples, yet adds compute to the critical path; spot GPUs offer 69–77\% lower cost, yet sit idle during training because DiT rollouts finish nearly simultaneously, which prevents LLM-style pipelining of rollout with training. Spot preemptions further break Sequence Parallelism (SP) groups, fragmenting GPU topology. We present Spotlight, the first system that harvests spot GPUs for DiT RL post-training. Spotlight rests on two key insights we devise: (1)~we show that exploration can tolerate stale model weights because exploration that uses the model weights from the previous iteration preserves the relative ranking of random seeds, allowing exploration to run on idle spot GPUs during training. (2)~SP reconfiguration can reuse on-node state, reducing group recovery from minutes to sub-second launches. Built on these insights, Spotlight introduces three techniques: a bandit-based exploration planner that maximizes reward variance within the training time budget, elastic sequence parallelism that reconfigures SP groups on the fly via persistent schedulers and intra-node weight copying, and a preemption-aware pull-based request scheduler that balances load and commits in-flight state upon preemption. We implement Spotlight on the open-source RL platform ROLL and evaluate it on Qwen-Image post-training. Spotlight reaches the same target validation score $4\times$ faster than baselines, reducing total cost by $1.4$-$6.4\times$ while achieving superior image quality on DeepSeek-OCR and Geneval datasets with resolution $512\times512$ and $1280\times1280$.

阅读与讨论 → 访问原文 →
20.
arXiv (CS.CL) 2026-06-17

In-Context Environments Induce Evaluation-Awareness in Language Models

作者:
Maheep Chaudhary↗

Humans often become more self-aware under threat, yet can lose self-awareness when absorbed in a task; we hypothesize that language models exhibit environment-dependent evaluation awareness. This raises concerns that models could strategically underperform, or sandbag, to avoid triggering capability-limiting interventions such as unlearning or shutdown. Prior work demonstrates sandbagging under hand-crafted prompts, but this underestimates the true vulnerability ceiling. We introduce a black-box adversarial optimization framework treating the in-context prompt as an optimizable environment, and develop two approaches to characterize sandbagging: (1) measuring whether models expressing intent to underperform can actually execute it across different task structures, and (2) causally isolating whether underperformance is driven by genuine evaluation-aware reasoning or shallow prompt-following. Evaluating Claude-3.5-Haiku, GPT-4o-mini, and Llama-3.3-70B across four benchmarks (Arithmetic, GSM8K, MMLU, and HumanEval), optimized prompts induce up to 94 percentage point (pp) degradation on arithmetic (GPT-4o-mini: 97.8\%$\rightarrow$4.0\%), far exceeding hand-crafted baselines which produce near-zero behavioral change. Code generation exhibits model-dependent resistance: Claude degrades only 0.6pp, while Llama's accuracy drops to 0\%. The intent – execution gap reveals a monotonic resistance ordering: Arithmetic $

阅读与讨论 → 访问原文 →
21.
arXiv (CS.LG) 2026-06-15

Graph Diffusion Residuals for Control-Function Instrumental Variables

作者:
Rui Wu↗Zongyuan Chen↗Hong Xie↗Defu Lian↗Enhong Chen↗

arXiv:2606.14636v1 Announce Type: new Abstract: Control-function instrumental variable estimators need a first-stage residual, not merely a first-stage prediction. High-capacity first stages can interpolate treatment and leave too little residual information for the outcome equation. We study Adaptive Anisotropic Instrumental Heat Flow (A-IHF), a deterministic graph-diffusion residual extractor for flexible control functions. A-IHF treats treatment as a signal on a graph of first-stage features, uses pilot diffusion to detect large treatment jumps, attenuates conductance across those jumps, and computes the generated control with a sparse graph resolvent. Its observational selection rule uses only $(Z,X)$, combining graph generalized cross-validation, roughness, residualized-treatment relevance, and graph-admissibility filtering. The analysis decomposes error into structural leakage, residual attenuation, and residualized treatment variation, yielding finite-sample bounds, graph-admissibility rates under latent piecewise-smooth geometry, and finite-path selection calibration. Across 54 synthetic benchmark cells with tuned graph, kernel, tree, boosting, series, and neural control-function baselines, guarded observational A-IHF has the lowest average structural-response MSE; the A-IHF family beats the best non-A-IHF baseline in 32 cells. Performance is strongest when the graph captures piecewise-smooth first-stage structure.

阅读与讨论 → 访问原文 →
22.
arXiv (CS.AI) 2026-06-16

UXBench: Measuring the Actionability of LLM-Generated UX Critiques

作者:
Wenjie Wang↗Yue Huang↗Zipeng Ling↗Han Bao↗Hang hua↗Xiaonan Luo↗Yu Jiang↗Shiyi Du↗Yuexing Hao↗Xiaomin Li↗Yuchen Ma↗Dianzhuo Wang↗…

arXiv:2606.16262v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as UX judges that inspect interfaces, diagnose usability problems, and propose repairs. Yet no controlled benchmark measures whether the resulting critiques are reliable and actionable across heterogeneous product surfaces. We introduce UXBench, a benchmark for evaluating LLMs as interaction-grounded UX judges. UXBench comprises local-first runnable web fixtures spanning ten product-surface families, paired with coverage-gated browser exploration that forces models to collect interaction evidence before reporting. Each judge model produces a structured UX report over seven rubric dimensions; report quality is measured by whether a fixed downstream repair agent can improve the interface based on the critique. We evaluate eight frontier models under both an automated repair-lift protocol and a blind human validation study. Results show that UX judging is neither saturated nor one dimensional: models differ meaningfully in report actionability, exhibit distinct rubric-level repair signatures, vary in fixture-level reliability, and trade leadership across surface categories

阅读与讨论 → 访问原文 →
23.
arXiv (CS.LG) 2026-06-15

Minimum Distance Summaries for Robust Neural Posterior Estimation

作者:
Sherman Khoo↗Dennis Prangle↗Song Liu↗Mark Beaumont↗

arXiv:2602.09161v2 Announce Type: replace-cross Abstract: Simulation-based inference (SBI) enables amortized Bayesian inference by first training a neural posterior estimator (NPE) on prior-simulator pairs, typically through low-dimensional summary statistics, which can then be cheaply reused for fast inference by querying it on new test observations. Because NPE is estimated under the training data distribution, it is susceptible to misspecification when observations deviate from the training distribution. Many robust SBI approaches address this by modifying NPE training or introducing error models, coupling robustness to the inference network and compromising amortization and modularity. We introduce minimum-distance summaries, a plug-in robust NPE method that adapts queried test-time summaries independently of the pretrained NPE. Leveraging the maximum mean discrepancy (MMD) as a distance between observed data and a summary-conditional predictive distribution, the adapted summary inherits strong robustness properties from the MMD. We demonstrate that the algorithm can be implemented efficiently with random Fourier feature approximations, yielding a lightweight, model-free test-time adaptation procedure. We provide theoretical guarantees for the robustness of our algorithm and empirically evaluate it on a range of synthetic and real-world tasks, demonstrating substantial robustness gains with minimal additional overhead.

阅读与讨论 → 访问原文 →
24.
bioRxiv (Bioinfo) 2026-06-12

DNA Compression with Genomic Language Models: Tokenization, Benchmarking, and an Information-Content Map

作者:
Macala↗Simecek↗

Lossless compression and probabilistic sequence modeling are two faces of the same coin: a model that assigns high probability to a sequence can encode it in few bits via arithmetic coding. We exploit this duality to evaluate genomic language models as compressors of DNA, using compression primarily as an objective probe of generative sequence modeling rather than as a deployable storage system. We release DNAGPT2, a family of ten GPT-2-small models pretrained for one epoch on a single A40 using the DNABERT2 multi-species corpus that differ only in byte-pair encoding vocabulary size. Coupled with arithmetic coding, the best model reaches 1.47 bits per base (bpb) on the T2T human genome, fourth in the Cobilab compression benchmark and ahead of every general-purpose compressor. Our results suggest that NLP-style tokenization choices may be suboptimal for DNA: a 32-token BPE vocabulary compresses better than larger vocabularies. We also find that, in this benchmark, published long-context genomic LMs underperform a much shorter-context BPE GPT-2; we discuss in Section 5 that this is not a controlled context-length ablation, since the compared models also differ in architecture, training data, parameter count, and tokenization. Finally, we compute a per-nucleotide information-content map of the human genome and show that exons, introns, intergenic regions, and Alu repeats have statistically distinct information profiles.

阅读与讨论 → 访问原文 →
25.
arXiv (CS.AI) 2026-06-11

Reinforcement Learning Disrupts Gradient-Based Adversarial Optimization

作者:
Xinhai Zou↗Chang Zhao↗Alireza Aghabagherloo↗Dave Singel\'ee↗Robin Degraeve↗Bart Preneel↗

arXiv:2606.12251v1 Announce Type: cross Abstract: Gradient-based adversarial attacks remain a dominant threat to deep neural networks (DNNs), as they exploit gradient information to efficiently optimize adversarial perturbations. To address this, we investigate whether reinforcement learning (RL) training can disrupt the gradient structure used by attackers by training image classifiers with policy-gradient objectives and epsilon-greedy exploration. Through systematic experiments across CIFAR-10, CIFAR-100, and ImageNet-100 with multiple architectures, we find that RL-trained classifiers significantly disrupt gradient-based adversarial optimization. To explain this, we conduct a comprehensive mechanism analysis using loss landscape visualization, static and dynamic gradient indicators, and predictive entropy. Our analysis reveals that RL acts as an implicit regularizer, producing models with highly unstable gradient directions and smaller gradient magnitudes. This combination makes each PGD step both unreliable in direction and limited in magnitude, causing gradient-based attacks to fail within practical iteration budgets. We further show that combining RL with adversarial training (RL-adv) provides a dual-layer defense operating at two complementary levels: RL degrades gradient information available to attackers (gradient-level defense), while adversarial training strengthens decision boundaries (boundary-level defense). RL-adv achieves the highest robustness across all major attack types evaluated, including gradient-based (PGD, AutoAttack), transfer-based, and query-based attacks, outperforming SL-adv by a significant margin. These findings identify RL-induced gradient disruption as a complementary robustness mechanism and motivate future research on hybrid SL-RL training schedules that combine SL's efficiency with RL's gradient-regularization properties.

阅读与讨论 → 访问原文 →