中 EN
Sign In Register

Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

Sign In Register
01.
medRxiv (Medicine) 2026-06-17

Impact of the disposable vape ban in Great Britain: a representative interrupted time-series study 2022-2026

Authors:
Tattan-Birch↗Jackson↗S. E↗Buss↗V. H↗Cox↗Shahab↗Kock↗Bauld↗Brown↗

Objective: To examine changes in vaping and smoking trends following the announcement and implementation of the disposable vape ban in Great Britain. Design: Interrupted time-series analysis of representative monthly cross-sectional data from the Smoking Toolkit Study. Setting: Great Britain. Participants: 118,946 adults ([≥]16y), including 12,042 young adults (16-24y), surveyed between Jan-2022 and Feb-2026. Main outcome measures: Changes in trends in disposable vape use among vapers, and current vaping and smoking prevalence, using seasonally-adjusted generalised additive models with comparisons against a no-ban counterfactual in which pre-announcement trends continued unchanged. Results: The proportion of vapers mainly using disposable devices began to decline following the announcement of the ban in Jan-2024, with the fall accelerating after implementation in June-2025. By Feb-2026, 5.6% (95%CI 4.6-6.9) of adult vapers and 7.1% (5.1-10.1) of young adult vapers mainly used disposables, compared with 62.0% (53.6-71.8) and 63.6% (52.7-76.7), respectively, under a no-ban counterfactual. Increases in vaping prevalence slowed post-announcement and plateaued post-implementation; by Feb-2026, prevalence was lower than the no-ban counterfactual in adults (13.6% v 18.8%; difference -5.2 percentage points, 95%CI -7.1 to -3.3) and young adults (27.8% v 39.1%; -11.3, -18.6 to -4.1). Declines in smoking prevalence stalled among adults and reversed among young adults post-announcement, before shifting downward again post-implementation; by Feb-2026, smoking prevalence was similar to the no-ban counterfactual in adults (difference +0.9 percentage points, -0.5 to +2.2) but possibly higher in young adults (+3.3, -0.5 to +7.1). Conclusions: The disposable vape ban in Great Britain was associated with substantial changes after both announcement and implementation, including a marked reduction in disposable vape use and a slowing then plateauing of growth in overall vaping prevalence. However, declines in smoking also temporarily slowed–and among young adults, reversed–after the announcement, before downward trends resumed after implementation.

Read & Discuss → View Source →
02.
arXiv (CS.AI) 2026-06-24

UniDrive: A Unified Vision-Language and Grounding Framework for Interpretable Risk Understanding in Autonomous Driving

Authors:
Xiaowei Gao↗Pengxiang Li↗Yitai Cheng↗Ruihan Xu↗James Haworth↗Stephen Law↗Yun Ye↗

arXiv:2606.24759v1 Announce Type: cross Abstract: Recent multimodal large language models (MLLMs) have shown strong potential for autonomous driving scene understanding, yet existing methods still face a fundamental trade-off between temporal reasoning and spatial precision. Models that rely on single-frame or low-resolution inputs often miss small, distant, or partially occluded hazards, while language-centric driving models frequently provide limited grounded evidence for their explanations. To address this gap, we propose UniDrive, a unified visual-language and grounding framework for interpretable risk understanding in autonomous driving. UniDrive combines a temporal reasoning branch that models scene dynamics from multi-frame visual input with a high-resolution perception branch that preserves fine-grained spatial details from the latest frame. The two branches are integrated through a gated cross-attention fusion module, enabling dynamic context to be aligned with precise spatial evidence. Based on the fused representation, UniDrive jointly generates natural-language risk descriptions and grounded bounding-box outputs for risk objects. Experiments on the DRAMA-Reasoning benchmark show that UniDrive outperforms representative image-based and video-based baselines in both captioning and risk-object grounding. In particular, UniDrive achieves the best overall performance on the validation split and demonstrates clear advantages in small-object localization, zero-shot generalization to NuScenes and BDD100K, and human-rated interpretability and trustworthiness. These results suggest that explicitly combining temporal semantics and high-resolution perception provides a stronger foundation for interpretable and safety-oriented autonomous driving systems. The code is available at https://github.com/pixeli99/unidrive-dev.

Read & Discuss → View Source →
03.
arXiv (CS.CL) 2026-06-24

Policies Permitting LLM Use for Polishing Peer Reviews Are Currently Not Enforceable

Authors:
Rounak Saha↗Gurusha Juneja↗Dayita Chaudhuri↗Naveeja Sajeevan↗Nihar B Shah↗Danish Pruthi↗

A number of scientific conferences and journals have recently enacted policies that prohibit LLM usage by peer reviewers, except for polishing, paraphrasing, and grammar correction of otherwise human-written reviews. But, are these policies enforceable? To answer this question, we assemble a dataset of peer reviews simulating multiple levels of human-AI collaboration, and evaluate five state-of-the-art detectors, including two commercial systems. Our analysis shows that all detectors misclassify a non-trivial fraction of LLM-polished reviews as AI-generated, thereby risking false accusations of academic misconduct. We further investigate whether peer-review-specific signals, including access to the paper manuscript and the constrained domain of scientific writing, can be leveraged to improve detection. While incorporating such signals yields measurable gains in some settings, we identify limitations in each approach and find that none meets the accuracy standards required for identifying AI use in peer reviews. Importantly, our results suggest that recent public estimates of AI use in peer reviews through the use of AI-text detectors should be interpreted with caution, as current detectors misclassify mixed reviews (collaborative human-AI outputs) as fully AI generated, potentially overstating the extent of policy violations.

Read & Discuss → View Source →
04.
arXiv (CS.CL) 2026-06-16

EIBench: A Simulator-Based Benchmark and Turn-Credit RL for Emotion Management

Authors:
Rongzhi Zhu↗Xiang Huang↗Yuchuan Wu↗Rui Wang↗Zequn Sun↗Tao Ren↗Weiyao Luo↗Bingxue Qiu↗Jieping Ye↗Yongbin Li↗Wei Hu↗

Emotional intelligence (EI) in Large Language Models (LLMs) is often evaluated through static understanding tasks or single-response dialogue generation. However, emotion management is interactive: a good model should not only recognize a user's emotion, but also improve the user's emotional and relational state over several turns. We introduce EIBench, a simulator-based benchmark for interactive emotion management. EIBench contains 2,222 scenarios, with 2,009 for training and 213 for held-out testing. The scenarios are organized by a 2x2 taxonomy covering Support, Defense, Repair, and Charm, which together capture different forms of support, boundary maintenance, trust repair, and rapport building. In each scenario, an LLM simulator plays the user, updates an emotion-relation state after each turn, and maps the final state to an anchor-based score. This design makes EIBench both an evaluation benchmark and a training environment: the final state gives the outcome reward, while the per-turn state updates provide dense feedback for RL. We evaluate 15 open- and closed-source LLMs. Current models perform well on support and rapport-building scenes, but struggle with boundary maintenance under user pressure. To improve the EI ability of LLMs, we propose Centered Turn-Credit GRPO (CTC-GRPO), a GRPO extension that reuses the simulator's per-turn state updates as dense turn-level feedback while preserving the final outcome reward. CTC-GRPO improves Qwen3-8B from -22.4 to +22.4 on EIBench and also improves on out-of-distribution evaluations including SAGE (+12.4) and EQBench3 (+20.9%). Our results show that simulator-tracked user states can support both evaluation and training for multi-turn emotion management.

Read & Discuss → View Source →
05.
arXiv (CS.AI) 2026-06-16

TNODEV: Toolbox for Neural ODE Verification

Authors:
Abdelrahman Sayed Sayed↗Pierre-Jean Meyer↗Mohamed Ghazel↗

arXiv:2606.16567v1 Announce Type: new Abstract: Neural ordinary differential equations (neural ODE) have started to appear in safety critical settings such as continuous-time controllers for cyber-physical systems and classifiers integrated into automated decision pipelines, raising the question of whether their behavior can be formally verified. Existing tools dedicated to neural ODE provide only a single reachability call without iterative input set refinement, limiting the precision of their verdicts to whatever one reachability call can deliver. We present TNODEV, the first sound formal verifier for neural ODE that integrates a falsification checker, a fast interval-based reachability backend based on continuous-time mixed monotonicity, a verification and refinement loop with three input-set splitting heuristics, and a parallel scheduler in a single end-to-end pipeline. TNODEV supports safe-set inclusion verification on pure neural ODE, neural ODE in closed loop with a neural network controller and general neural ODE (GNODE), with the safe set specified either as an interval or as the half-space intersection induced by a target classification label. We evaluate TNODEV on a range of benchmarks across safe-set inclusion and classification-robustness properties, including a direct reachability comparison against NNV~2.0 and CORA and a verification comparison against NNV2.0 on MNIST general neural ODE classifiers.

Read & Discuss → View Source →
06.
arXiv (CS.AI) 2026-06-18

TLA-Prover: Verifiable TLA+ Specification Synthesis via Preference-Optimized Low-Rank Adaptation

Authors:
Eric Spencer↗Arslan Bisharat↗Brian Ortiz↗Khushboo Bhadauria↗TaiNing Wang↗George K. Thiruvathukal↗Konstantin Laufer↗Mohammed Abuhamad↗

arXiv:2606.06133v2 Announce Type: replace-cross Abstract: TLA+ is a formal specification language for verifying distributed systems and safety-critical protocols. Large language models (LLMs) frequently produce TLA+ specifications that fail the TLC model checker for semantic reasons. Across 25 LLMs, the best public baseline is 26.6% syntactic parse and 8.6% semantic model-check. We present TLA-Prover, a 20-billion-parameter model for TLA+ specification synthesis. Training combines supervised fine-tuning (SFT) on verified examples with repair-based group-relative policy optimization (GRPO). In the GRPO stage, the model learns to fix its own rejected specifications. We also train a direct preference optimization (DPO) variant from the same SFT checkpoint as an ablation. TLC provides the reward signal directly, with no learned reward model. Four tiers grade each output: Bronze (parses), Silver (no warnings), Gold (passes TLC), and Diamond. To reach Diamond, the model's correctness property is automatically altered in a small way; TLC must then detect a violation. If TLC still passes, the property was always-true and contributes nothing; the output fails Diamond. TLA-Prover reaches 9/30 (i.e. pass@1 = 30%) at both Gold and Diamond on a held-out 30-problem benchmark. This is roughly 3.5x the 8.6% untuned baseline. The DPO variant reaches 20% at Diamond. Gold and Diamond coincide at every checkpoint; this prevents the trivial-property failure mode.

Read & Discuss → View Source →
07.
arXiv (CS.AI) 2026-06-11

Engineering Robustness into Personal Agents with the AI Workflow Store

Authors:
Roxana Geambasu↗Mariana Raykova↗Pierre Tholoniat↗Trishita Tiwari↗Lillian Tsai↗Wen Zhang↗

arXiv:2605.10907v3 Announce Type: replace-cross Abstract: The dominant paradigm for AI agents is an "on-the-fly" loop in which agents synthesize plans and execute actions within seconds or minutes in response to user prompts. We argue that this paradigm short-circuits disciplined software engineering (SE) processes – iterative design, rigorous testing, adversarial evaluation, staged deployment, and more – that have delivered the (relatively) reliable and secure systems we use today. By focusing on rapid, real-time synthesis, are AI agents effectively delivering users improvised prototypes rather than systems fit for high-stakes scenarios in which users may unwittingly apply them? This paper argues for the need to integrate rigorous SE processes into the agentic loop to produce production-grade, hardened, and deterministically-constrained agent *workflows* that substantially outperform the potentially brittle and vulnerable results of on-the-fly synthesis. Doing so may require extra compute and time, and if so, we must amortize the cost of rigor through reuse across a broad user community. We envision an *AI Workflow Store* that consists of hardened and reusable workflows that agents can invoke with far greater reliability and security than improvised tool chains. We outline the research challenges of this vision, which stem from a broader flexibility-robustness tension that we argue requires moving beyond the ``on-the-fly'' paradigm to navigate effectively.

Read & Discuss → View Source →
09.
arXiv (CS.LG) 2026-06-12

From Uncertain Judgments to Calibrated Rankings: Conformal Elo Estimation for LLM Evaluation

Authors:
Bora Kargi↗David Salinas↗

arXiv:2606.13221v1 Announce Type: new Abstract: Evaluating new large language models typically requires costly human annotation campaigns at scale. LLM-as-a-judge offers a cheaper alternative, but judge scores carry systematic errors - such as position bias, self-preference, or intransitivity - that can strongly miscalibrate the resulting rankings. We quantify the resulting judge-human disagreement at two complementary levels. At the local level, we estimate per-battle uncertainty from the judge's own score differences by propagating calibrated win probabilities rather than hard labels into the Bradley-Terry procedure. This alone provides a drastic improvement to Elo estimation accuracy, bringing LLM-derived ratings within 17.9 Elo MAE of human-derived ones when averaged over 55 held-out models on LMArena. At the global level, we apply split conformal prediction to the residual gap between LLM-derived and human-derived Elo ratings across held-out models, producing prediction intervals with distribution-free marginal coverage guarantees that account for irreducible LLM-human disagreement. Together, these two layers yield a low-cost evaluation tool that provides developers with calibrated Elo estimates and honest uncertainty bounds, without access to large-scale human annotations.To facilitate reproducibility, we release our code at https://github.com/kargibora/SoftElo .

Read & Discuss → View Source →
10.
medRxiv (Medicine) 2026-06-15

GLLaucoMed: A Secure LLM-Powered Agentic Workflow for Automated Medication Extraction from Free-Text Glaucoma Clinical Notes

Authors:
Solages↗Scherer↗Samico↗G. A↗Gutkind↗N. E↗Kang↗Medeiros↗F. A↗Swaminathan↗S. S↗

Purpose: To evaluate the efficacy of large language models (LLMs) in extracting medication-related information from glaucoma clinical notes in the electronic health record (EHR). Design: Cross-sectional. Subjects: 1,250 subjects in the Bascom Palmer Ophthalmic Repository. Methods: Extracted clinical notes from glaucoma-related encounters between 2014 and 2024 were labeled by two glaucoma specialists with a third serving as an adjudicator. Graders were asked to label current topical medications (CTM), proposed changes to topical medications ({Delta}TM), current oral medications (COM), and proposed changes to oral medications ({Delta}OM) in a structured fashion. The dataset was split into development (10%), validation (10%), and test (80%) sets stratified by clinician. Development and validation sets were used to engineer and refine prompts, and the held-out test set was used for model assessment. Five LLMs (Claude Opus 4.6, DeepSeek-V3.2, GPT 5.2, Grok 4.1, and Qwen3.6-35B-A3B) were accessed via Microsoft Azure AI Foundry within a HIPAA-compliant environment. Inter-grader agreement was assessed with Gwet AC1. LLM performance was initially assessed in a binary fashion with F1 scores, and the degree of text match among positive cases was evaluated using exact match accuracy and Jaccard Index (JI). Main Outcome Measures: F1 score, exact match accuracy, JI. Results: Gwet AC1 for intergrader agreement was 0.799, 0.888, 0.985, and 0.988 for CTM, {Delta}TM, COM, and {Delta}OM, respectively. F1 scores for CTM were 0.985, 0.971, 0.978, 0.968, and 0.970 for Claude, Deepseek, GPT, Grok, and Qwen, respectively; for {Delta}TM: 0.905, 0.826, 0.897, 0.842, 0.855, respectively; for COM: 0.923, 0.887, 0.899, 0.906, 0.894, respectively; for {Delta}OM: 0.958, 0.815, 0.937, 0.835, 0.940, respectively. Among positive cases, range of exact match accuracies for CTM (N=1354) was 0.730- 0.882 and range of JIs was 0.809-0.918. For {Delta}TM (N=404), exact match accuracy range was 0.619-0.780 and JI range was 0.668-0.827. For COM (N=47), exact match accuracy range was 0.766-0.872 and JI range was 0.765-0.870. For {Delta}OM (N=25), exact match accuracy range was 0.583-0.920 and JI range was 0.583-0.922. Conclusions: The GLLaucoMed pipeline demonstrated high performance in extracting and standardizing medication data from unstructured clinical notes, including both current medications and proposed changes. Claude and GPT exhibited the strongest performance.

Read & Discuss → View Source →
11.
arXiv (CS.CL) 2026-06-16

The Value Axis: Language Models Encode Whether They're on the Right Track

Authors:
Nick Jiang↗Isaac Kauvar↗Jack Lindsey↗

We investigate whether language models internally track the value of their current trajectory, defined as the likelihood that their ongoing strategy will achieve their goals. Using synthetic, in-context reinforcement learning data, we construct a "value" axis for Qwen3-8B. We find that activations along this axis distinguish between high vs. low verbalized confidence, rollouts without and with backtracking, and correct vs. corrupted code. Steering towards high value causally suppresses self-correction and reduces explanatory verbosity, while steering towards low value induces backtracking and exploration. We demonstrate that direct preference optimization (DPO) can increase the internal value of rewarded behaviors (e.g. use a certain word), causing the model to act more confidently after exhibiting them. Finally, we apply the value axis to study in-the-wild settings. For example, we find that Qwen assigns low value to politically sensitive chat queries after post-training and that supervised fine-tuning increases internal confidence within the training domain. Our results suggest that language models linearly encode an estimate of expected goal success that modulates their confidence in pursuing a direction.

Read & Discuss → View Source →
12.
arXiv (CS.CL) 2026-06-16

Understanding the Behaviors of Environment-aware Information Retrieval

Authors:
Ruifeng Yuan↗Chaohao Yuan↗David Dai↗Yu Rong↗Hong Cheng↗Hou Pong Chan↗Chenghao Xiao↗

Recent retrieval-augmented generation (RAG) approaches have demonstrated strong capability in handling complex queries, yet current research overlooks a critical challenge: different retrievers require fundamentally different query formulation strategies for optimal performance. In this work, we present the first systematic analysis of how LLMs can learn to adapt their query formulation strategies for different retrievers via reinforcement learning (RL). Our empirical study reveals that RL effectively teaches an LLM to tailor its queries to specific retriever characteristics. We discover that different retrievers exhibit surprisingly distinct optimal query styles (e.g., descriptive vs. question-like), suggesting strategies learned for one retriever ineffective for another. We further show that performance can be enhanced by incorporating retriever-specific human guidance and by scaling model size. To facilitate learning over multi-retrieval-step trajectories, we introduce a branching-based rollout technique that improves training stability. Our work provides the first empirical evidence and actionable insights for building truly retriever-aware RAG systems. Code and resources are available at https://github.com/LCO-Embedding/Envs-aware-Information-Retrieval.

Read & Discuss → View Source →
13.
arXiv (math.PR) 2026-06-24

Genealogical processes of sequential Monte Carlo methods and other non-neutral population models under rapid mutation

Authors:
Jere Koskela↗Paul A. Jenkins↗Adam M. Johansen↗Dario Span\`o↗

arXiv:2406.16465v3 Announce Type: replace Abstract: We show that genealogical trees arising from a broad class of non-neutral models of population evolution converge to the Kingman coalescent under a suitable rescaling of time. As well as non-neutral biological evolution, our results apply to genetic algorithms encompassing the prominent class of sequential Monte Carlo (SMC) methods. The time rescaling we need differs slightly from that used in classical results for convergence to the Kingman coalescent, which has implications for the performance of different resampling schemes in SMC algorithms. In addition, our work substantially simplifies earlier proofs of convergence to the Kingman coalescent, and corrects an error common to several earlier results.

Read & Discuss → View Source →
14.
arXiv (CS.CV) 2026-06-11

Frozen Foundation-Model Embeddings Discard Small-Lesion Signal in Chest Radiography: Implications for Pre-Deployment Evaluation

Authors:
Raajitha Muthyala↗Zhenan Yin↗Alekhya Jilla↗Frank Li↗Theo Dapamede↗Bardia Khosravi↗Mohammadreza Chavoshi↗Judy Gichoya↗Saptarshi Purkayastha↗

Frozen vision-transformer (ViT) foundation-model embeddings increasingly serve as the substrate for downstream chest-radiography (CXR) pipelines, yet where small-scale, low-contrast signal is retained or lost in the frozen forward pass has not been systematically quantified across architectures, pretraining domains, and objectives. We probed five frozen ViTs (RAD-DINO, DINOv2-B/14, DINOv3 ViT-7B, BiomedCLIP, MedSigLIP) and a frozen DINO-pretrained ResNet-50 architectural control across three large CXR cohorts (NIH-CXR14, MIMIC-CXR, Emory-CXR; aggregate pool n=492,724) and ChestX-Det10 (n=3,543; 1,462 small-lesion bounding boxes across Calcification, Nodule, Mass). Each model was evaluated with a small-scale-perturbation panel and a region-aware bounding-box-stratified probe on real lesions, comparing three pooling modes from the same forward pass: classification token (CLS), patch-mean (mean over all final-layer patch tokens), and bounding-box-restricted patch-local. On the perturbation panel, CLS embeddings sat at the chance floor (area under the ROC curve [AUC] 0.500-0.524); patch-mean was indistinguishable from CLS on iso-blur and reticular-fine cells but rose with CLS on larger directional-blur footprints, while disease AUC on globally decided tasks ranged 0.642-0.913. Patch-local probes recovered AUC ~1.0 from the same forward pass (per-model mean improvement +0.412 to +0.488); the ResNet-50 control reproduced the chance floor. On ChestX-Det10, image-level CLS classification showed within-class small-versus-large stratum gaps up to +0.243 AUC; bounding-box-level patch-local pooling on the same forward pass recovered AUC >= 0.899 on every (model x class) cell. Frozen ViT embeddings silently suppress small-scale signal at the global-aggregation step; the signal is recoverable from patch tokens conditional on a region of interest.

Read & Discuss → View Source →
15.
arXiv (CS.LG) 2026-06-18

Unraveling the Mechanism of Drug Binding to SARS-CoV-2 RNA Pseudoknot with Thermodynamics-Driven Machine Learning

Authors:
Mariia Ivonina↗Jakub Rydzewski↗

arXiv:2604.14906v3 Announce Type: replace-cross Abstract: The pseudoknot secondary structure in SARS-CoV-2 RNA is essential for regulating protein synthesis through $-$1 programmed ribosomal frameshifting ($-1$ PRF), a mechanism that allows the virus to generate both structural and non-structural proteins from overlapping reading frames. This pseudoknot exhibits both threaded and unthreaded long-lived topologies. The influence of ligand binding on its folding is a process critical for the development of $-$1 PRF small-molecule inhibitors. Understanding this process through unbiased molecular dynamics (MD) simulations can be facilitated by introducing collective variables (CVs) that capture the corresponding slowest dynamical modes. Here, we use spectral map (SM), a thermodynamics-driven machine learning technique, to learn such CVs directly from all-atom MD trajectories of the SARS-CoV-2 RNA pseudoknot in complex with the $-$1 PRF inhibitor merafloxacin and its two structural analogs in neutral and ionized forms. Free-energy landscapes (FELs) derived from the learned CVs indicate that ligand-induced destabilization is topology-selective. In the threaded pseudoknot, the inhibitors destabilize the S2 stem, while in the unthreaded pseudoknot, destabilization occurs in the S1 and S3 stems. Furthermore, the extent to which each ligand reshapes the FEL matches experimentally reported antiviral potency, whereas the protonation state qualitatively alters dynamics within the same RNA topology. Overall, our results show how pseudoknot topology, ligand type, and protonation state collectively influence the slow conformational dynamics of viral RNA and establish physiological protonation as a critical factor for modeling RNA-targeted drug action.

Read & Discuss → View Source →
16.
arXiv (CS.LG) 2026-06-19

BLISS: A Lightweight Bilevel Influence Scoring Method for Data Selection in Language Model Pretraining

Authors:
Jie Hao↗Rui Yu↗Wei Zhang↗Huixia Wang↗Jie Xu↗Mingrui Liu↗

arXiv:2510.06048v5 Announce Type: replace Abstract: Effective data selection is essential for pretraining large language models (LLMs), enhancing efficiency and improving generalization to downstream tasks. However, existing approaches often require leveraging external pretrained models, making it difficult to disentangle the effects of data selection from those of the external pretrained models. In addition, they often overlook the long-term impact of selected data if the model is trained to convergence, primarily due to the prohibitive cost of full-scale LLM pretraining. In this paper, we introduce BLISS (BileveL Influence Scoring method for data Selection): a lightweight data selection method that operates entirely from scratch, without relying on any external pretrained oracle models, while explicitly accounting for the long-term impact of selected data. BLISS leverages a small proxy model as a surrogate for the LLM and employs a score model to estimate the long-term influence of training samples if the proxy model is trained to convergence. We formulate data selection as a bilevel optimization problem, where the upper-level objective optimizes the score model to assign importance weights to training samples, ensuring that minimizing the lower-level objective (i.e., training the proxy model over the weighted training loss until convergence) leads to best validation performance. Once optimized, the trained score model predicts influence scores for the dataset, enabling efficient selection of high-quality samples for LLM pretraining. We validate BLISS by pretraining 410M/1B/2.8B Pythia and LLaMA-0.5B models on selected subsets of the C4 dataset. Notably, under the 1B model setting, BLISS achieves $1.7\times$ speedup in reaching the same performance as the state-of-the-art method, demonstrating superior performance across multiple downstream tasks.

Read & Discuss → View Source →
18.
arXiv (CS.AI) 2026-06-15

When Good Verifiers Go Bad: Self-Improving VLMs Can Regress on New Tasks

Authors:
Jianzhe Lin↗

arXiv:2606.14629v1 Announce Type: cross Abstract: Verifier-driven self-DPO is a common recipe for self-improving production visual-language models. In this setup, a frozen verifier scores candidate generations, the top- and bottom-scoring candidates form a preference example, and DPO updates the learner. The deployment-time assumption is monotone: a stronger verifier should yield a stronger student. We show that this assumption can fail because verifier quality is highly task-specific. On a four-rung open-source verifier ladder across MathVista, MMMU, and BLINK, the same verifiers that are above-threshold and improve a Qwen-3-VL-2B student on MathVista become sub-threshold on MMMU, where their task-rubric accuracy drops to 8% to 23%. In this regime, every verifier we tested silently regresses the student, producing drops of 3.4 to 10.9 percentage points below the frozen baseline while the DPO training loss continues to decrease. The regression replicates on a second student, Qwen-2.5-VL-3B. Moreover, within the failure regime, damage is confidence-inverted: the more accurate-but-still-wrong verifier causes larger regression than a near-random verifier, suggesting that progress-gated replay amplifies confidently wrong preference pairs. We give a compact mechanistic explanation via a variance theorem for progress-gated replay and its direction-mismatch failure mode. The deployment message is operational rather than purely diagnostic: before running any verifier-driven loop, teams should measure target-task rubric accuracy, rank verifiers by target-task rubric quality rather than parameter count, and treat diminishing returns in above-threshold regimes as a verifier-side compute budget cap.

Read & Discuss → View Source →
19.
arXiv (CS.AI) 2026-06-15

CSPO: Constraint-Sensitive Policy Optimization for Safe Reinforcement Learning

Authors:
Ayoub Belouadah↗Sylvain Kubler↗Yves Le Traon↗

arXiv:2606.14415v1 Announce Type: new Abstract: Safe reinforcement learning (Safe RL) aims to maximize expected return while satisfying safety constraints, typically modeled as Constrained Markov Decision Processes (CMDPs). While primal-dual methods scale well to deep RL, they often suffer from delayed constraint correction, leading to oscillatory behavior and prolonged safety violations. In this paper, we propose Constraint-Sensitive Policy Optimization (CSPO), a first-order primal-dual method that incorporates local constraint sensitivity into policy updates. CSPO augments the primal objective with a constraint-sensitive correction derived from the shortest signed distance to the safety boundary, enabling smarter recovery steps back to safety, compensating for delayed Lagrange multiplier updates, reducing oscillations near the boundary, and preserving the KKT solutions of the original constrained problem. Experiments on navigation and locomotion benchmarks demonstrate that CSPO achieves faster safety recovery and high reward preservation, resulting in higher constrained returns compared to state-of-the-art primal-dual and penalty-based methods

Read & Discuss → View Source →
20.
arXiv (CS.AI) 2026-06-16

Safe Exploration via Policy Priors

Authors:
Manuel Wendl↗Yarden As↗Manish Prajapat↗Anton Pollak↗Stelian Coros↗Andreas Krause↗

arXiv:2601.19612v3 Announce Type: replace-cross Abstract: Safe exploration is a key requirement for reinforcement learning (RL) agents to learn and adapt online, beyond controlled (e.g. simulated) environments. In this work, we tackle this challenge by utilizing suboptimal yet conservative policies (e.g., obtained from offline data or simulators) as priors. Our approach, SOOPER, uses probabilistic dynamics models to optimistically explore, yet pessimistically fall back to the conservative policy prior if needed. We prove that SOOPER guarantees safety throughout learning, and establish convergence to an optimal policy by bounding its cumulative regret. Extensive experiments on key safe RL benchmarks and real-world hardware demonstrate that SOOPER is scalable, outperforms the state-of-the-art and validate our theoretical guarantees in practice.

Read & Discuss → View Source →
21.
arXiv (CS.AI) 2026-06-11

Implicit Neural Representations of Individual Behavior

Authors:
Andrew Kang↗Priya Narasimhan↗

arXiv:2606.12200v1 Announce Type: cross Abstract: We study policy representation learning from unlabeled multi-policy behavioral data. Each episode is generated by a fixed policy, but policy labels are unavailable. This setting appears in robotics play, demonstrations, games, racing, and other datasets where heterogeneous behaviors are mixed without annotations. We introduce Behavioral INR, a self-supervised generative model that adapts implicit neural representations (INRs) from vision to behavior. Instead of mapping coordinates to RGB values, Behavioral INR represents a policy as a state-action function mapping states to subsequent actions. An episode-level latent modulates this function through FiLM layers, yielding a generative prior over policies and allowing policy identity to be inferred without supervision. Because INRs treat each datapoint as samples from an underlying function, the same model naturally accommodates variable episode lengths and different sampling granularities, as in vision INRs with different image resolutions. We also define policy-level out-of-distribution (OOD) shifts along state-distribution and action-distribution axes, which arise when policies overlap in states or actions but are not captured by standard behavioral OOD settings based only on new agents or environments. We evaluate on synthetic Gaussian random field data, MuJoCo demonstrations with controlled OOD splits, and real-world chess, Formula 1 racing, robotics, and Seek-Avoid datasets. Behavioral INR most consistently improves policy identifiability in the hardest continuous state-action settings, especially when longer episodes, more policies, and OOD splits reduce the usefulness of marginal shortcuts; amortized history encoders remain competitive when policy identity can be recovered from symbolic repetition or low-dimensional action statistics. We release code and checkpoints.

Read & Discuss → View Source →
22.
arXiv (CS.LG) 2026-06-24

Forget Without Compromise: Nexus Sampling for Streaming KV-Cache Eviction Under Fixed Budgets

Authors:
Duc Duong↗Hoang Anh Duy Le↗Jianwen Xie↗Anshumali Shrivastava↗Zhaozhuo Xu↗

arXiv:2606.23961v1 Announce Type: new Abstract: Long-context and agentic LLM workloads push the KV cache past any fixed memory budget, forcing the inference stack to permanently evict tokens at every step of a continuous-inference stream. Existing methods all share the same template, a per-step direct-attention score followed by deterministic top-$K$ selection, which converts a single below-cutoff step into an irreversible verdict and permanently erases any subtly important token that direct attention cannot single out from noise. To address this challenge, we propose Nexus Sampling, a training-free eviction method that pairs Nexus scoring, an iterative walk over direct attention that surfaces bridge tokens, with weighted reservoir sampling, which retains tokens with inclusion probability in place of deterministic top-$K$. Theoretically, we show that Nexus Sampling dominates deterministic top-$K$ in long-run survival of subtly important tokens. Empirically, at 80% KV cache eviction, Nexus Sampling matches dense attention within 1% on LongBench while outperforming top-$K$ baselines on retrieval-heavy tasks, with up to 10x smaller per-sequence cache memory.

Read & Discuss → View Source →
23.
arXiv (CS.LG) 2026-06-19

A Model-Driven Approach for Developing Families of Reinforcement Learning Environments

Authors:
Xiaoran Liu↗Istvan David↗

arXiv:2606.20324v1 Announce Type: cross Abstract: Virtual training environments are software-intensive systems in which reinforcement learning (RL) agents learn, adapt, and demonstrate meaningful behavior. Virtual training environments offer a safe and cost-efficient alternative to training agents in real-world settings. However, to converge, most realistic RL problems require training in multiple, mostly similar but slightly different environments - i.e., families of environment variants. The typical development process of environment families is a labor-intensive and error-prone manual endeavor that does not scale well. To alleviate these issues, in this paper, we propose a model-driven approach for developing families of RL training environments. To obtain the family of environments, we develop an approach and prototype tool. In our approach, a hybrid genetic algorithm - a combination of population-based global search and heuristic local search - generates environment families. Mutations and constraints are expressed as model transformations and are operationalized into a search process by a state-of-the-art model transformation engine. We demonstrate the soundness of our approach in a wildfire mitigation scenario and curriculum learning - a particular learning paradigm that relies on environment families.

Read & Discuss → View Source →
24.
arXiv (CS.CL) 2026-06-17

Teaching Values to Machines: Simulating Human-Like Behavior in LLMs

Authors:
Asaf Yehudai↗Naama Rozen↗Ariel Gera↗

Large Language Models (LLMs) demonstrate a remarkable capacity to adopt different personas and roles; however, it remains unclear whether they can manifest behavior that adheres to a coherent, human-like value structure. In this work, we draw on established psychological value theory to induce human-like values in LLMs and assess their alignment with patterns observed in human studies. Using validated psychological questionnaires, we conduct large-scale experiments – over 5 million questions – to evaluate value structures and value-behavior relationships in leading LLMs and compare them to humans. Our findings reveal strong agreement between value-prompted LLMs and humans across both dimensions. Moreover, incorporating human value distributions enhances population-level simulations with value-induced LLMs. These findings highlight the potential of value-induced LLMs as effective, psychologically grounded tools for simulating human behavior.

Read & Discuss → View Source →
25.
arXiv (CS.AI) 2026-06-18

Caring Without Feeling: Affective Dynamics as the Control Layer of Human-AI Agent Collaboration

Authors:
Junjie Xu↗Xingjiao Wu↗Zihao Zhang↗Yujia Xu↗Yuzhe Yang↗Jin Zhu↗Luwei Xiao↗Wen Wu↗Liang He↗

arXiv:2606.18259v1 Announce Type: cross Abstract: AI agents that plan, retain memory across sessions, invoke external tools and act with partial autonomy are transforming human–AI collaboration. Research on affective computing, simulated empathy in large language models, trust in automation and AI safety has illuminated important design principles, yet these literatures remain fragmented. No integrated account explains how affective cues operate within agentic collaboration – settings in which humans delegate, monitor and correct consequential tasks. This Review synthesises computational and interactional mechanisms of affective dynamics: the processes through which affective cues, emotion-like behaviour and perceived agent affect shape trust calibration, delegation decisions, error correction, dependence and governance. We trace how model-generated affective signals enter interaction loops that govern reliance, repair and oversight, and propose a framework that treats affect not as an internal property of AI but as a coordination layer through which humans and agents negotiate capability, uncertainty and responsibility. The framework provides a foundation for calibrated measurement, purposeful design and informed governance.

Read & Discuss → View Source →