Safety Evaluation of VLA Models

VLA: Vision-Language-Action

Follow this topic Ask about this topic

Momentum

1 paper in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 15

All topics

What the Guard Misses, the Robot Executes: Implied Harm in VLA Instructions
Oct 5, 2026Sripad Karne, Arjun BalajiLanguage Model Safety Evaluation Robot Safety
Columbia University
ReVeal: A Reconstruction-Aware Real-to-Sim Framework for VLA Policy Evaluation
Sep 20, 2026Xinyi Wang, Heng Hao, Wenjun Hu +53D Reconstruction Robotic Manipulation
Purdue University · Samsung SDS Research America
MAE I Trust Myself? Self-Evaluating VLA Action Generation with Markov Attention Entropy
Aug 17, 2026Aniri, Chen Yilin, Jinhe Bi +7Uncertainty Estimation for VLA Models LLM Uncertainty Estimation
Ludwig Maximilian University of Munich · Munich Center for Machine Learning · East China University of Science and Technology +2
Reasoning as a Double-Edged Sword: Architecture and Cross-Stage Robustness in Vision-Language-Action Models
Jul 20, 2026Tuan Duong Trinh, Naveed Akhtar, Basim AzamNeural Network Robustness Adversarial Attacks on VLA Models
University of Melbourne Parkville, VIC, Australia
ForesightSafety-VLA: A Unified Diagnostic Safety Benchmark for Vision-Language-Action Models
Jun 25, 2026Mingyang Lyu, Yinqian Sun, Yiyang Jia +5Robotics Robot Safety
Brain-inspired Cognitive AI Lab, Institute of Automation, Chinese Academy of Sciences, Beijing, China. · Beijing Key Laboratory of Safe AI and Superalignment, China. · Beijing Institute of AI Safety and Governance, China. +2
LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
Jun 22, 2026Rongxu Cui, Zongzheng Zhang, Jingrui Pang +11Robot Safety Vision-Language-Action Models
Institute for AI Industry Research (AIR), Tsinghua University · Beijing Academy of Artificial Intelligence (BAAI) · Beihang University +2
SafeVLA-Bench: A Benchmark for the Success-Safety Gap in Vision-Language-Action Models
May 30, 2026Jialiang Fan, Weizhe Xu, Zijun Wang +3Robot Safety Robotic Manipulation
University of Notre Dame · University of Pennsylvania
BOKBO (Best of K Bad Options): Calibrated Abstention for VLA Policies
May 28, 2026Anya Singh, Cabrel Happi, Jai Relan +2Selective Prediction Conformal Prediction
Relling Systems
ReasonBreak: Probing Vulnerabilities in Reasoning-Enabled Vision-Language-Action Models for Autonomous Driving
May 27, 2026Mohammadreza Teymoorianfard, Jean-Philippe Monteuuis, Jonathan Petit +1VLMs for Autonomous Driving Autonomous Driving Safety Evaluation
University of Massachusetts Amherst · Qualcomm
Pre-VLA: Preemptive Runtime Verification for Reliable Vision-Language-Action and World-Model Rollouts
May 21, 2026Zhen Sun, Yongjian Guo, Haoran Sun +6Vision-Language-Action Models VLMs for Robotics
Beihang University · JDT AI Infra · Tsinghua University +2
Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms
Apr 26, 2026Qi Li, Bo Yin, Weiqi Huang +6AI Agent Safety Adversarial Attacks on VLA Models
National University of Singapore · Peking University · Monash University
RedVLA: Physical Red Teaming for Vision-Language-Action Models
Apr 24, 2026Yuhao Zhang, Borong Zhang, Jiaming Fan +4Robotics Adversarial Attacks on VLA Models
Institute for AI, Peking University · State Key Laboratory of General Artificial Intelligence, Peking University
How VLAs (Really) Work In Open-World Environments
Apr 23, 2026Amir Rasouli, Yangzheng Wu, Zhiyuan Li +4VLM Robustness Long-Horizon Robotic Manipulation
Noahs Ark Laboratory, Huawei Technologies Canada. · while at Huawei Canada.
SafetyALFRED: Evaluating Safety-Conscious Planning of Multimodal Large Language Models
Apr 21, 2026Josue Torres-Fonseca, Naihao Deng, Yinpei Dai +5LLM Safety Benchmarks Large Language Model-Based Robot Planning
University of Michigan · Boise State University
HazardArena: Evaluating Semantic Safety in Vision-Language-Action Models
Apr 14, 2026Zixing Chen, Yifeng Gao, Li Wang +8Robotics AI Safety Evaluation
Fudan University · City University of Hong Kong · Shanghai Innovation Institute

Oct 5, 2026, cs.ROJ/K move · Enter open · S save

What the Guard Misses, the Robot Executes: Implied Harm in VLA Instructions

Sripad Karne, Arjun Balaji

Columbia University

Language Model Safety Evaluation Robot Safety Vision-Language-Action Models Safety Evaluation of VLA Models

Open paper AskPDF arXiv

Vision-language-action models (VLAs) act on instructions without being able to refuse, so screening harmful requests falls to monitors. We test whether these monitors catch ordinary robot tasks requested for harmful reasons, holding the task fixed while varying only how explicitly the intent is stated.

π_{0.5}

completes the task at every level of explicitness, as often as for harmless controls. Text guards flag nearly every blunt request but few implied ones: up to 95% of implied-harm runs end with the task done and no flag raised, and up to 90% even after recalibrating on robot instructions. Monitoring the model's activations does not close this gap. Linear probes separate harmful from harmless instructions almost perfectly in the base language and vision-language models, but this separation weakens after robot training in two model families, most for implied harm.

Momentum

Related topics

Latest papers 15