safety

PAVXploreRL: Physical-Action-Visual World Model Reinforcement Learning with Action Exploration

arXiv:2607.16602v2 Announce Type: replace Abstract: Action-conditioned world models are a key component of embodied AI, serving as scalable policy evaluators that reduce reliance on expensive real-wor

Safety

PGTT: Phase-Guided Terrain Traversal for Perceptive Legged Locomotion

DGX agent

arXiv:2510.18348v2 Announce Type: replace-cross Abstract: State-of-the-art perceptive Reinforcement Learning controllers for legged robots typically either (i) impose oscillator-or IK-based gait prior

safetyarxiv-cs-ai

23 Jul 2026

Predictive Extrema, Unprofitable Policies: An AI-Assisted Audit of Candle-Based Binance Spot Timing Models

arXiv:2607.19453v1 Announce Type: cross Abstract: We audit whether candle-based machine-learning models can turn predictions of cryptocurrency extrema or short-horizon outcomes into positive Binance S

Privileged Lesion-Context Relational Distillation for Mask-Free Skin Lesion Classification

arXiv:2607.18773v1 Announce Type: new Abstract: Accurate skin lesion classification can benefit from lesion segmentation masks, but requiring masks or an auxiliary segmentation model during inference

Prompt Programming for Cultural Bias and Alignment of Large Language Models

arXiv:2603.16827v2 Announce Type: replace Abstract: Culture shapes reasoning, values, prioritization, and strategic decision-making, yet large language models (LLMs) often exhibit cultural biases that

PyroDash: Cost-Efficient Token-Level Small-Large Language Model Collaborative Inference

arXiv:2607.20327v1 Announce Type: new Abstract: Large language models (LLMs) provide strong reasoning capabilities but are expensive to serve at scale, whereas small language models (SLMs) are cheaper

Rater State Bias in RLHF Preference Data: An Audit Framework

arXiv:2607.16195v2 Announce Type: replace Abstract: We identify a structured confound in Reinforcement Learning from Human Feedback (RLHF). Pairwise preference labels are intended to reflect the compa

Really insightful Substack post from @GaryMarcus ref the recent OpenAI -HuggingFace incident. It provided much need context around initial m…

Really insightful Substack post from @GaryMarcus ref the recent OpenAI -HuggingFace incident. It provided much need context around initial media reporting. Two points especially stood out to me: #Tech

safetygary-marcus--x

REGEN: Replay-recycling for Expert-to-Generalist distillation with Offline Reinforcement Learning

arXiv:2607.19450v1 Announce Type: cross Abstract: Large-scale online reinforcement learning (RL) is the predominant means of eliciting advanced abilities including long-term reasoning and agentic tool

Rewarding Better Thinking for LLM Preference Alignment

arXiv:2607.19824v1 Announce Type: new Abstract: LLM preference alignment aims to optimize models toward human preferences across diverse user instructions. Reinforcement learning has become a major po

SafeGen: Goal-Conditioned Video Diffusion of Safety-Critical Scenarios for VLM-Based Autonomous Driving

arXiv:2607.19701v1 Announce Type: new Abstract: VLMs are increasingly deployed in AD systems, creating an urgent need for rigorous safety evaluation under rare yet safety-critical scenarios. Among the

Scale-Aware Learning of Chaotic Dynamics on Unstructured Meshes via Binned Spectral Losses

arXiv:2607.19387v1 Announce Type: cross Abstract: Surrogate modeling for high-dimensional nonlinear dynamical systems that exhibit chaos requires mechanisms that preserve not only pointwise accuracy b

SEED: Towards More Accurate Semantic Evaluation for Visual Brain Decoding

arXiv:2503.06437v3 Announce Type: replace-cross Abstract: We present SEED (Semantic Evaluation for Visual Brain Decoding), a novel metric for evaluating the semantic decoding performance of visual bra

SeededGrasp: Language-Guided Grasping in Complex Scenes with Multiple Embodiments

arXiv:2607.20207v1 Announce Type: new Abstract: Practical robotic grasping in complex scenes requires both 3D spatial reasoning and alignment with task-specific requirements. Vision-language models (V

safetyarxiv-cs-ro

Self-Explaining Reinforcement Learning for Mobile Network Resource Allocation

arXiv:2509.14925v2 Announce Type: replace Abstract: Deep reinforcement learning (DRL) methods, though powerful, often lack transparency, which limits their adoption in critical domains. We apply Self-

SemICP: Semantic Non-Rigid Point Cloud Registration with Elastic Energy Regularization

arXiv:2503.00972v4 Announce Type: replace Abstract: Purpose: Accurate point cloud registration is essential in computer-aided interventions (CAI) to align multi-modal medical images for intraoperative

Simulating Eutopia: Revisiting Long-term Fairness with Outcomes, Performativity, and Dynamics

arXiv:2607.19389v1 Announce Type: cross Abstract: As AI-driven Decision Makers (ADMs) influence our socioeconomic reality, their roles in both enhancing efficiency and amplifying the social biases hav

SLPO: Scaling Latent Reasoning via a Surrogate Policy

arXiv:2607.19691v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards has become the predominant recipe for eliciting test-time scaling in explicit Chain-of-Thought reasoner

SOPD-SocialNav: Selective On-Policy Distillation for Vision-Language Social Navigation

arXiv:2607.19850v1 Announce Type: new Abstract: Vision-language models have shown strong potential for social robot navigation by leveraging rich semantic understanding of complex environments and hum

safetyarxiv-cs-ro

Sound Probabilistic Safety Bounds for Large Language Models

arXiv:2607.20286v1 Announce Type: cross Abstract: We propose a novel framework for computing rigorous bounds on the probability that a large language model (LLM) generates harmful output to a given pr

Spatiotemporal Facial Action Unit Detection using Twin Cycle Autoencoders for Driver Monitoring

arXiv:2607.16760v2 Announce Type: replace-cross Abstract: Driver monitoring systems (DMS) increasingly rely on facial cues to infer drowsiness, distraction, and cognitive load in real time. Facial Act

Stale but Stable: Staleness-Adaptive Trust Regions for Stabilizing Asynchronous Reinforcement Learning

arXiv:2607.18722v2 Announce Type: replace-cross Abstract: Asynchronous reinforcement learning improves throughput by decoupling rollout generation from optimization, but staleness is an inevitable byp

Statevector-Referenced Geometry Survival of a Four-Qubit ZZ Quantum Kernel on IBM Quantum Hardware: A Fixed-Subset Diagnostic Across Three Execution Configurations

arXiv:2607.20377v1 Announce Type: cross Abstract: Quantum-kernel methods encode a dataset's geometry in a Gram matrix, so learning claims on hardware kernels assume the intended geometry survives exec

STEREOFLOW: Progressive Stereo Matching with StereoDiT and Transition Flow Matching

arXiv:2607.19986v1 Announce Type: new Abstract: Stereo matching is a fundamental task in 3D reconstruction. Despite remarkable advances, the prevailing paradigms formulate stereo matching as a determi

Stochastic Primal-Dual Decoding for Multiobjective Generative Recommender Systems

arXiv:2607.19357v1 Announce Type: new Abstract: Recent advances in recommender systems (RS) have shown substantial performance gains through generative modelling. In practice, recommendation often inv

StreamHOI: Interaction-aware Temporal Memory Adaptation for Streaming HOI Video Generation

arXiv:2607.20174v1 Announce Type: cross Abstract: Existing human--object interaction (HOI) video generation methods are largely limited to offline short-video generation with complex driving condition

Stress Testing Concept Erasure with Large Language Model Agents

arXiv:2607.17890v2 Announce Type: replace Abstract: Concept erasure aims to remove semantic concepts from a trained generative model and is increasingly important for responsible AI deployment. Howeve

Structured Latent Space Modeling over Multi-Scale Temporal Patches for Multivariate Time Series Forecasting

arXiv:2607.19404v1 Announce Type: cross Abstract: Multivariate time series encode structural patterns that unfold across multiple temporal scales, yet most forecasting backbones treat learned represen

Style over Substance: A Shortcut Audit of Emotion-Description Preference Evaluation

arXiv:2607.18508v1 Announce Type: new Abstract: Preference over model-generated emotion descriptions is emerging as a standard evaluation metric for multimodal emotion understanding, exemplified by th

TalentCLEF at CLEF2026: Skill and Job Title Intelligence for Human Capital Management

arXiv:2607.20009v1 Announce Type: new Abstract: This paper presents the second edition of the TalentCLEF Challenge, which will run as an evaluation lab as part of CLEF 2026. The aim of TalentCLEF is t

TAP-RAG: Task-Aware Policy Control for Long-Document Multimodal Question Answering

arXiv:2607.18917v1 Announce Type: new Abstract: Long-document multimodal question answering requires more than retrieving relevant chunks from a large document. Different queries require different evi

Test Case Prioritization for DNNs via Neural Collapse Instability

arXiv:2607.20046v1 Announce Type: cross Abstract: With the widespread deployment of deep neural networks (DNNs) in safety-critical domains, reducing the cost of model validation under limited testing

Test-Time Registers as Global Priors for Tokenized Image Generation

arXiv:2607.16824v2 Announce Type: replace Abstract: Attention-based models often develop attention sinks, where a small number of tokens repeatedly attract attention and accumulate unusually large act

The Ethics of Autonomous AI Agents for Offensive Security

arXiv:2607.20255v1 Announce Type: cross Abstract: LLM-driven autonomous agents are reshaping offensive security. Unlike traditional penetration-testing tooling -- deterministic, narrowly scoped, and o

The Geometry of Learning to Avoid Interventions

arXiv:2602.03825v2 Announce Type: replace Abstract: Human interventions are a common source of supervision in autonomous systems during deployment. Many existing approaches are based on avoiding inter

The Mechanism Matters: When Knowledge Graphs Help Reinforcement Learning

arXiv:2607.19616v1 Announce Type: new Abstract: Knowledge graphs (KGs) are widely used to inject prior knowledge into reinforcement learning (RL), yet the literature is dominated by single-domain, pos

The Two-Process Theory of Machine Self-Report

arXiv:2607.20082v1 Announce Type: new Abstract: Language models are increasingly asked to self-report, informing safety evaluations, public understanding, and model-welfare debates. Yet their reports

Thoughtful essay on employment from @Anthropic’s @PeterMcCrory, endorsed by Google DeepMind’s @alexolegimas. I tend also agree with most of …

Thoughtful essay on employment from @Anthropic’s @PeterMcCrory, endorsed by Google DeepMind’s @alexolegimas. I tend also agree with most of it, which is why I have consistently suggested that a jobapo

safetygary-marcus--x

Toward Adaptable Multi-Agent Reinforcement Learning: An Assumption-Aware Review

arXiv:2507.10142v2 Announce Type: replace Abstract: Multi-Agent Reinforcement Learning (MARL) has achieved strong performance in simulated benchmarks, yet real deployments often violate the assumption

Towards Torque-Driven Reinforcement Learning for Quadruped Locomotion

arXiv:2607.18365v1 Announce Type: cross Abstract: Reinforcement learning (RL) for legged robots is advancing locomotion, demonstrating its ability to adapt to new and challenging terrain. Traditionall

TRUST-ESD: A Risk-Calibrated and Governance-Aware AI Framework for Enterprise Strategic Decision Support Under Uncertainty

arXiv:2607.20065v1 Announce Type: new Abstract: Enterprise strategic decision support requires AI systems that are not only accurate, but also uncertainty-aware, risk-calibrated, explainable, and gove

Trustworthy Privacy-Preserving Multimodal Federated Learning for Personalised Breast Cancer Prediction

arXiv:2607.19532v1 Announce Type: cross Abstract: Federated learning has emerged as a potential solution to privacy concerns associated with using sensitive health data for training predictive models,

Variance-reduced Domain Adaptation using Paired Sampling

arXiv:2607.20367v1 Announce Type: new Abstract: Correlation alignment and the maximum mean discrepancy are two widely used distribution-matching frameworks for unsupervised domain adaptation (UDA). Ho

Wave2Body: Rethinking mmWave Human Pose Estimation as Radar-to-Body Token Translation

arXiv:2607.18875v1 Announce Type: new Abstract: Millimeter-wave (mmWave) radar enables privacy-friendly human sensing, but its sparse point clouds are physical measurements of view-dependent electroma

WearWow: Native 2K Multi-Garment Virtual Try-On via Adaptive Token Packing and Preference Alignment

arXiv:2607.19923v1 Announce Type: new Abstract: Synthesizing native 2K multi-garment virtual try-on is a formidable frontier in digital fashion, critically bottlenecked by two fundamental limitations:

What Matters in Humanoid General Motion Tracking? An Empirical Study

arXiv:2607.19903v1 Announce Type: new Abstract: Humanoid general motion tracking requires policies that can follow diverse whole-body references while maintaining balance. Building such policies invol

safetyarxiv-cs-ro

When Reasoning Narrows the Move: Diversity Collapse in LLM Game Play

arXiv:2607.19523v1 Announce Type: new Abstract: Supervised fine-tuning (SFT) is widely used to adapt large language models to downstream tasks, but its effect on behavioral diversity in sequential dec

Which Values Do LLMs Confuse? A Schwartz-Based Recognition Study

arXiv:2607.20270v1 Announce Type: new Abstract: Large language models are increasingly evaluated through the values they endorse, but such evaluations presuppose that models can identify the value exp