Safety Filtering

Ask about this topic

Overview

Safety filtering studies how to detect and prevent unsafe behavior at runtime or during training, from robot collisions to harmful language-model outputs and risky tool calls. In control, common approaches use control barrier functions, quadratic-program action shields, reachability and backup policies, with learned models and conformal calibration addressing uncertainty. For language and multimodal systems, filters include trained guard classifiers, hidden-state probes, streaming checks, and pre-execution tool monitors. A central challenge is balancing reliable protection against unnecessary intervention, especially when attacks, environments, or user contexts differ from those seen in evaluation.

Latest papers

CardsList

126 papers14 papers in the last four weeks, down 39% on the four weeks before. 0.1% of all new papers.

  1. Not an Accident: Exploiting Safety Filter Inadequacy in Humanoid Robots

    Oct 22, 2026Tiffany Bao, Shangtao Li, Yifan Sun +6Robot SafetySafety Filtering

    Arizona State University · Carnegie Mellon University

  2. SITO: Sifting a Commercial Text-Safety API for Missed Harm and False Flags in English, Ukrainian and Surzhyk

    Oct 19, 2026Andrii TuzhykovContent ModerationSafety Filtering

    AI/ML Consultant / PhD Researcher The State Institution South Ukrainian National Pedagogical University named after K. D. Ushynsky

  3. Safe Reinforcement Learning with Receding-Horizon Safety Critic and Differentiable Action Shields

    Oct 14, 2026Siwei Ju, Jan Peters, Oleg ArenzSafe RLSafety Filtering

    Department of Computer Science, Technical University of Darmstadt, Germany. · Hessian.AI. · German Research Center for AI (DFKI), Research Department: Systems AI for Robot Learning. +1

  4. Jacobian State Augmented Viable Kinematic Safety Filter

    Oct 11, 2026Siddharth Singh, Joshua VaughanSafety FilteringRobotic Control

    Manufacturing Robotics and Controls Group within the Manufacturing Science Division, Energy Science & Technology Directorate of the Oak Ridge National Lab.

  5. CSF: Contextual Safety Filtering for Motion Generators

    Oct 8, 2026Lizhi Yang, Yiling Hou, Yao Tang +4Safety FilteringRobot Safety

    California Institute of Technology, CA 91125, USA · New York University, NY 10012, USA

  6. Context-Conditioned Hamilton-Jacobi Reachability for Adaptive Safety Filtering

    Oct 6, 2026Ali Fuat Sahin, Yunus Yazoglu, John Talbot +3Reachability AnalysisAutonomous Driving Safety Evaluation

    Ecole Polytechnique Fédérale de Lausanne · Stanford University · Toyota Research Institute

  7. Reading, Not Manipulating: Leveraging Router Logits for Multimodal Safety in MoE Vision-Language Models

    Oct 6, 2026Ziyuan Yang, Wenxuan Ding, Shangbin Feng +1Vision-Language ModelsSafety Filtering

    University of Washington · New York University

  8. A Safe Prototype Is Not a Safety Direction: Reference Dependence and Prompt Confounds in Response-Safety Embeddings

    Oct 1, 2026Sahil KadadekarPrototype-Based ClassificationLLM Evaluation

    New York University

  9. Multi-Link Safety Filtering for VLA Policies Around Moving Hazards

    Sep 30, 2026Yatharth Agarwal, Vijay RaghunathanRobot SafetySafety Filtering

    School of Electrical and Computer Engineering Purdue University, West Lafayette, IN, USA

  10. Guard Models Are Overconfident Where Base Models Are Uncertain

    Sep 29, 2026Jonghyun Hong, MinJae Jung, Minwoo KimLanguage Model Safety EvaluationAdversarial Attacks on LLMs

    DATUMO INC.

  11. Adaptive Safety Filtering for Frozen ACC Policies via Conformal Residual Calibration

    Sep 28, 2026Zhiruo Zhou, Rigaudiere Z. Li, Chen Xiwen +3Safety-Critical ControlAutonomous Driving Safety Evaluation

    Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China · Wuhan University of Technology, Wuhan, China · Shanghai Jiao Tong University, Shanghai, China

  12. Predictive Semantic Safety: From Visual Physical Reasoning to Safety-Critical Control

    Sep 28, 2026Taekyung Kim, Salem Fradi, Yanning Dai +2Robotic ControlSafety-Critical Control

    Department of Robotics, University of Michigan, Ann Arbor, MI 48109, USA · Center of Excellence in Generative AI, King Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia. · Dalle Molle Institute for Artificial Intelligence Research (IDSIA), Switzerland. +2

  13. When World Models Lie: Adaptive Safety Analysis Under Wrong Imaginations

    Sep 28, 2026John Cao, Somil BansalReachability AnalysisWorld Models for Robotics

    Department of Aeronautics and Astronautics, Stanford University

  14. AdaGuard: An Adaptive Guard Model with User-defined Policies

    Sep 28, 2026Yunhao Feng, Yifan Ding, Yuxiang Xie +4LLM GuardrailsAI Agent Safety

    1National University of Defense Technology · 2Fudan University

  15. COGNIT-Guard: Calibrated Standalone Direct-Decision Guardrails with Heterogeneous CPU-NPU Confidence Cascading under Explicit Latency and False-Positive Constraints

    Sep 27, 2026Hao ChenLLM GuardrailsLanguage Model Calibration

    School of Cyber Security, Guangdong Polytechnic Normal University, Guangzhou, China

  16. Safety Reconstructed: Generative Modeling via Masked Diffusion Builds Strong Safety Guardrails

    Sep 27, 2026Gert Lek, Abele Malan, Chaoyi Zhu +3LLM Safety BenchmarksLLM Guardrails

    University of Neuchâtel · Delft University of Technology · IBM Research +1

  17. Baszta: Data-Centric Fine-Tuning of a Polish Multi-Label Safety Classifier

    Sep 24, 2026Adam Górski, Mateusz Jąkalak, Rafał JakubowskiMulti-Label ClassificationLanguage Model Safety Evaluation

    Billennium S.A.

  18. CrossSafe: Towards Cross-Embodiment Latent Safety Filters

    Sep 24, 2026Ihab Tabbara, Yuxuan Yang, Hussein SibaiReachability AnalysisCross-Embodiment Robot Learning

    Department of Computer Science and Engineering, Washington University in St. Louis, St. Louis, MO, USA

  19. LEAP-CBF: A Safety Filter for Uncertain Systems with Least-Effort Adversarial Potentials

    Sep 23, 2026Oswin So, Eric Yu, Chuchu FanSafety-Critical ControlControl Barrier Functions

    Massachusetts Institute of Technology.

  20. InGuard: Towards Generalized Inner Guardrail for Safe Text-to-Image Generation

    Sep 23, 2026Zeyu Wang, Xiaodan Li, Zhiwen Li +2T2I GenerationSafe T2I Generation

    Alibaba AAIG

  21. Safety-Filtered Distributed Koopman-MPC

    Sep 23, 2026Shengjun Zhang, Wenhao Li, Zhenxin Lin +1Model Predictive ControlRobot Collision Avoidance

    School of Artificial Intelligence, Hubei University, Wuhan, Hubei, 430061 China · Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS), Shenzhen, Guangdong, China · School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), Shenzhen, Guangdong, China +1

  22. Turning Safety into Competence: Minimally Exploitable Robot Policies via Safety-Filtered Reinforcement Learning

    Sep 23, 2026Ruihan Wu, Rui Yang, Donggeon David Oh +2Sequential MARLRL for Robotics

    Department of Computer Science, Johns Hopkins University, Baltimore, MD 21218, USA · Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ 08544, USA

  23. Can Data Attribution Filter Out Subliminal Learning? Not Reliably

    Sep 17, 2026Moritz Weckbecker, Sweta Jena, Jonas Müller +5Gradient-Based AttributionTraining Data Curation

    Department of Artificial Intelligence, Fraunhofer Heinrich Hertz Institute · International Institute of Information Technology Hyderabad · Department of Electrical Engineering and Computer Science, Technische Universität Berlin +4

  24. Runtime Safety Filtering for Two-Terminal Hazards in Robotic Battery Recycling

    Sep 17, 2026Yuxin Cao, Wei Song, Xianglin Yang +4Robot SafetySafety Filtering

    National University of Singapore, Singapore · Griffith University, Australia · UNSW Canberra, Australia +2

  25. Safety Beyond the Interface: Detecting Harm via Latent States in Large Language Models

    Sep 16, 2026Alizishaan Khatri, Chiquita Prabhu, Omkar NeogiSafety FilteringLLM Safety

    Wrynx Inc. · Independent Researcher

  26. Winning a Won Game: Strict Reach-Avoid-Stay Control Barrier Functions for High-Dimensional Black-Box Systems

    Sep 16, 2026Donggeon David Oh, Duy P. Nguyen, Gongkai Yuan +3Robotic ControlControl Barrier Functions

    Department of Computer Science, Johns Hopkins University, Baltimore, MD 21218, USA

  27. Timely Activation of Safety Filters via One-Step Reachability Expansion

    Sep 15, 2026Javier BorquezReachability AnalysisSafety-Critical Control

    Universidad de Santiago de Chile Santiago, Chile

  28. ResSafe: Learning Safety Filtering with Residual Reinforcement Learning for Humanoids

    Sep 14, 2026Gechen Qu, Tong Zhang, Bike Zhang +4Safety-Critical ControlHumanoid Robot Control

    University of California, Berkeley · University of California, Los Angeles

  29. Exact Feasibility Certification and Optimal Responsibility Allocation for Multi-Robot CBF Safety Filters

    Sep 14, 2026Chandan Kumar Sah, Jishnu KeshavanSafety-Critical ControlControl Barrier Functions

    Department of Mechanical Engineering, Indian Institute of Science, Bengaluru 560012, India

  30. Certified Safety Curation: Distribution-Free Guarantees for Safe Offline Reinforcement Learning

    Sep 10, 2026Adam Haroon, Cody FlemingOffline RLSafety Filtering

    Iowa State University Ames, IA, USA