safety

RELATE: An Evaluation Framework for measuring Relational Orientation of Large Language Models

arXiv:2610.09569v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for emotional support, raising concern that sustained use may draw users away from their real-world r

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Rephrase Before You Act: Characterizing and Mitigating Language Sensitivity in Vision-Language-Action Models

DGX agent

arXiv:2610.10526v1 Announce Type: cross Abstract: Vision-language-action models (VLAs) are strikingly sensitive to instruction phrasing and do not inherit the language robustness of the vision-languag

safetyarxiv-cs-cl

8 Oct 2026

Revisiting Explainable AI through Model-Independent Concept Dictionaries

arXiv:2610.10301v1 Announce Type: new Abstract: Modern applications of AI rely on increasingly complex models. Explainable AI (XAI) has emerged as a set of techniques aimed at improving model transpar

RFPO: Rectified Flow Policy Optimization for Embodied Control

arXiv:2610.10453v1 Announce Type: new Abstract: Flow-based policies provide an expressive framework for continuous robot control, but their iterative ODE integration incurs substantial inference cost.

RoboPace: Contact-Aware Time-Optimal Retiming for Action-Chunk Policies

arXiv:2610.09696v1 Announce Type: new Abstract: Robot manipulation data collection has been shifting from teleoperation toward robot-free demonstrations, through interfaces such as the Universal Manip

RoboPrompt: Intuitive Robot Policy Steering with Sparse Human Input

arXiv:2610.10534v1 Announce Type: new Abstract: End-to-end robot policies trained through imitation learning remain constrained by limited data diversity, making reliable zero-shot deployment in real-

RoboRender: Robot-Oriented Video Generation for Visual Sim-to-Real Transfer

arXiv:2610.09254v1 Announce Type: new Abstract: Simulation enables large-scale, low-cost robot data generation, but policies trained in simulation often fail to transfer to the real world due to the s

RollVerify: Bridging Efficiency and Accuracy in Long-Tail Rollout Reinforcement Learning

arXiv:2610.09914v1 Announce Type: new Abstract: Reinforcement learning is crucial for improving large language models' reasoning and generalization. It relies on massive rollouts whose lengths become

Rubix: Global Correspondence-Free Point Set Alignment through Assignment Geometry

arXiv:2610.10408v1 Announce Type: new Abstract: Procrustes-Wasserstein alignment jointly estimates a matching and rotation without supplied correspondences, but alternating minimization can stop at su

Safe Meta-Policy Design with Risk Control

arXiv:2610.10393v1 Announce Type: cross Abstract: Models can be retrained as new data arrive, but deploying every new version risks replacing a good policy with a worse one. We study how to plan polic

Safe on Average, Unsafe in the Tail: When Is the Episodic-Cost Tail Controllable?

arXiv:2610.09508v1 Announce Type: new Abstract: Safe reinforcement learning seeks policies that maximize return while satisfying constraints on cumulative cost. Most methods impose these constraints o

SAPD: Step-Aligned Privileged Distillation

arXiv:2610.09665v1 Announce Type: new Abstract: On-policy post-training can improve large language models by learning from their own trajectories, but requires costly rollout generation. We ask whethe

SAPS: Shared Autonomy for Policy Steering by Blending Teleoperation with a Pretrained VLA

arXiv:2606.15568v2 Announce Type: replace Abstract: Vision-Language-Action (VLA) models perform well in generalist robot manipulation but remain brittle to out-of-distribution perturbations, while ful

Scalable Logistic Gaussian Process Density Regression with Kinetic Langevin Sampling

arXiv:2610.09591v1 Announce Type: cross Abstract: Conditional density estimation targets the full distribution of a response given covariates, as required, for example, for per-galaxy photometric reds

SearchWorld: Spatial Value-Grounded Imagination for UAV Object Search via World Models

arXiv:2610.09335v1 Announce Type: cross Abstract: Autonomous unmanned aerial vehicle (UAV) object search involves a closed loop of perception, decision-making, and action under partial observability.

SeOPD: Self-Evolving LLMs via Online Policy Distillation from Self-Generated Chain-of-Thought

arXiv:2609.33181v2 Announce Type: replace-cross Abstract: Recent advances in online policy self-distillation (OPSD) have demonstrated that large language models (LLMs) can improve their capabilities b

SGF+: Decoupling Gradient Flows for Autoregressive Video Generation

arXiv:2610.10429v1 Announce Type: new Abstract: Autoregressive video generation requires denoising the current frames while writing their key-value representations as context for future predictions. H

Share GPU clusters across teams with isolation and fairness using Amazon SageMaker HyperPod

A reference architecture for securely sharing one Amazon SageMaker HyperPod EKS cluster across multiple teams, using AWS IAM Identity Center for authentication, per-team SageMaker Domains and Kubernet

safetyaws-ml-blog

Shared Gaussianization: What Gaussian Regularizers Certify About Contrastive Learning, and What They Miss

arXiv:2610.10299v1 Announce Type: new Abstract: What can a distribution-matching regularizer such as SIGReg in LeJEPA certify about contrastive learning? We study shared Gaussianization (SG), a charac

Shared Geometry As A Rosetta Stone: Cross-Modal Alignment Without Paired Data

arXiv:2610.09411v1 Announce Type: cross Abstract: Multimodal representations enable zero-shot classification and retrieval, but aligning independently trained models usually requires large amounts of

Shared-Roadmap Generation and Evaluator for Multi-Agent Path Planning Using Heterogeneous Graph Neural Network

arXiv:2610.09034v1 Announce Type: cross Abstract: Multi-agent path planning (MAPP) in continuous environments often relies on roadmaps to balance safety and search efficiency. However, traditional roa

SkillCycle: Co-Evolving Agent Policies and Skill Banks

arXiv:2610.09430v1 Announce Type: new Abstract: Internalizing external skills changes a language agent's capabilities and, with them, the value of its remaining guidance: rules can become redundant, m

SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning

arXiv:2602.08234v2 Announce Type: replace Abstract: Large Language Model (LLM) agents have shown stunning results in complex tasks, yet they often operate in isolation, failing to learn from past expe

SOTA: Stock Options Trading Agents Guided by Option-Implied Return Distributions

arXiv:2610.10407v1 Announce Type: cross Abstract: As option markets grow and AI advances, agentic systems for option trading are gaining increasing attention. Language-model-based agents can reason ov

Sparse Feature Policy Unlearning Mitigates State Hallucination in Vision-Language-Action Models

arXiv:2610.09496v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have shown strong generalization in robotic manipulation by leveraging rich representations from pretrained vision

Spatial Induction Heads: In-Context Learning of Multidimensional Cellular Automata

arXiv:2610.09124v1 Announce Type: new Abstract: Induction heads provide a mechanistic account of in-context learning in sequential data, but existing theory largely assumes that the context relevant t

Stationary Bias and Extrapolation in Nonlinear Two-Timescale Stochastic Approximation

arXiv:2610.10246v1 Announce Type: new Abstract: Constant-step stochastic approximation generally has a nonzero stationary mean error that persists under time averaging. This paper studies that error f

Symmetry-Informed Causal Partial Identification

arXiv:2610.09230v1 Announce Type: new Abstract: Partial identification (PI) entails estimating bounds on causal effects by encoding different assumptions on data generation as a constrained optimizati

Taming an End-to-End Autonomous Driving Policy for Urban Navigation of Quadruped Robots

arXiv:2610.08812v1 Announce Type: new Abstract: We present Go2-DrivoR, a goal-conditioned adaptation of the end-to-end autonomous driving trajectory planning framework DrivoR for urban navigation with

TAP: Efficient Long-Horizon Agent Pruning via Trajectory-Anchored Recovery

arXiv:2610.09074v1 Announce Type: new Abstract: Emerging long-horizon agentic tasks require repeated model calls, worsening the inference cost of already-costly language models. While narrow agentic t

Targeted Modality Dropout for Real-Robot Manipulation Robust to Intermittent Vision Loss

arXiv:2610.09566v1 Announce Type: new Abstract: Imitation learning policies that integrate multiple sensory modalities are prone to overreliance on a dominant modality, such as vision, during training

TempoBridge: Language-Guided Tempo Control for Vision-Language-Action Policies

arXiv:2610.09451v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models are effective at understanding what task to perform, but provide limited control over how it should be executed, suc

TERRA: Learning Transportable Latent Actions through Temporal Effect Representation and Relational Alignment

arXiv:2610.09509v1 Announce Type: new Abstract: Latent actions supervise robot policies with action-like codes inferred from visual transitions, and their usefulness hinges on two questions: what a co

The Deceptive Bandit Problem: Exploratory Coupling and the Fragility of Multi-Agent Learning

arXiv:2610.09120v1 Announce Type: new Abstract: Randomized exploration is central to bandit learning, multi-agent reinforcement learning, and zeroth-order policy search, yet its independence and priva

The Symbol of the Surrogate: Measuring Numerical Provenance in Neural PDE Solvers

arXiv:2610.09255v1 Announce Type: new Abstract: Neural PDE surrogates are trained on numerical solver outputs that contain both physical evolution and solver-specific discretization errors. Because su

TiTok: Audio-Visual LLM for Multi-Segment Temporal Grounding

arXiv:2610.09408v1 Announce Type: new Abstract: Audio-visual multi-segment grounding (AV-MSG) in untrimmed videos, reasoning over audio-visual evidence and predicting multiple segments for a query, is

TMT: Runtime Backdoor Detection for Vision-Language-Action Policies on Unseen Tasks

arXiv:2610.09462v1 Announce Type: new Abstract: Backdoored vision-language-action (VLA) policies can preserve benign task performance while producing malicious actions when a trigger appears. Detectin

TouchScale: 500 Hours of Human Vision and Touch for Visual-Tactile Learning

arXiv:2610.10288v1 Announce Type: new Abstract: Large-scale egocentric human interaction data is becoming an important source of physical supervision for embodied learning, yet video alone leaves the

Toward Evidence-Driven Human-Agent-Robot Teaming for Earth-Independent Anomaly Triage

arXiv:2610.08933v1 Announce Type: new Abstract: Deep-space crews cannot rely on real-time ground support for urgent off-nominal events. Initial alerts may underdetermine cause, while discriminating ev

🧬 Training and evaluating AI scientists on drug target discovery is hard. We introduce DrugTargetWorld: Synthetic Biomedical Worlds for Tra…

🧬 Training and evaluating AI scientists on drug target discovery is hard. We introduce DrugTargetWorld: Synthetic Biomedical Worlds for Training and Benchmarking AI Scientists The environment turns en

safetygary-marcus--x

Transition Path Sampling Using Koopman Operators and Exit-Time Optimal Control

arXiv:2610.10054v1 Announce Type: cross Abstract: Sampling transitions between metastable states is a central problem in dynamical systems theory and molecular dynamics in particular. A key challenge

Two-Level Softmax Sampling Done Right: Correcting Bias from Size Imbalance and Dispersion

arXiv:2610.10483v1 Announce Type: new Abstract: Sampling from a softmax distribution is a fundamental operation in machine learning, but its linear complexity in the number of items makes exact sampli

Understanding and Mitigating Token-Pruning-Induced Vulnerabilities in VLMs

arXiv:2610.09703v1 Announce Type: cross Abstract: Token-Pruning accelerates Vision-Language Models by removing redundant visual tokens, yet its safety implications remain underexplored. In this work,

Unpaired Canonical Correlation Analysis

arXiv:2610.09530v1 Announce Type: cross Abstract: Canonical Correlation Analysis (CCA) is a fundamental method for multiview shared space learning. However, its strict reliance on paired data poses a

Using Small Language Models to Reverse-Engineer Machine Learning Pipelines Structures

arXiv:2610.10261v1 Announce Type: cross Abstract: Context: Once defined a taxonomy of stages structuring Machine Learning (ML) pipelines (e.g. Data Preprocessing, Modeling...), extracting these stages

Validity Without Ground Truth: What Stated-Preference Economics Offers the Evaluation of Language Models

arXiv:2610.10506v1 Announce Type: cross Abstract: Many of the questions now put to large language models have no correct answer to score against: what a policy is worth, which option a user should cho

We have a choice in front of us. In my latest op-ed for Transformer, I invite employees of frontier AI companies to ask themselves if their …

We have a choice in front of us. In my latest op-ed for Transformer, I invite employees of frontier AI companies to ask themselves if their work is truly leading us toward a safer future. Reflecting o

safetyyoshua-bengio--x

Workhorse: Learning Robust Whole-Body Humanoid Loco-Manipulation from Human Data

arXiv:2610.09117v1 Announce Type: cross Abstract: Humanoid robots still struggle to plan contact-rich whole-body manipulation from egocentric RGB and proprioception. Workhorse learns such manipulation