Safety Filtering

Follow this topic Ask about this topic

Momentum

24 papers in the last four weeks, up 118% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 111

All topics

kNNGuard: Turning LLM Hidden Activations into a Training-Free Configurable Guardrail
Jul 2, 2026Mahmoud Abdelfattah, Hamid Nasiri, Peter GarraghanLLM Guardrails Safety Filtering
Lancaster University · Mindgard
FastBridge: Closing the Model-Based Realization Gap in Safety Filters on 3D Gaussian Splatting for Fast Quadrotor Flight
Jul 1, 2026Tscholl Dario, Nakka Yashwanth Kumar, Gunter BrianCollision Avoidance 3D Gaussian Splatting
Safe Alone, Unsafe Together: Safeguarding Against Implicit Toxicity When Benign Images Combine
Jul 1, 2026Jiaxian Lv, Shiyao Cui, Yingkang Wang +3Toxicity Detection Safety Filtering
The Conversational AI (CoAI) Group, DCST, Tsinghua University
Do Safety Guardrails Need to Reason? LeanGuard: A Fast and Light Approach for Robust Moderation
Jun 25, 2026Dongbin NaLLM Guardrails CoT Reasoning
POSTECH
Safe responses matter: Output-aware safety guardrail mitigate over-refusal in MLLMs
Jun 20, 2026Jiayi Li, Kun ZhanLLM Guardrails Multimodal Large Language Models
Lanzhou University
OTTER: A Red-Teaming System for Toxicity-Evading Jailbreak Prompt Optimization
Jun 19, 2026Jerry Wang, Hsin-Ling Hsu, Yi-Cheng Lai +2Adversarial Prompt Generation Adversarial Attacks on LLMs
1National Chengchi University, Taipei, Taiwan
OSGuard: A Benchmark for Safety in Computer-Use Agents
Jun 13, 2026Mina Mohammadmirzaei, Jeffrey FlaniganComputer-Use Agents LLM Guardrails
University of California, Santa Cruz
Provably Safe, Yet Scalable Reinforcement Learning
Jun 12, 2026Kai S. Yun, Zeyang Li, Navid AzizanRL Control Safety-Critical Control
MIT
Stop Early, Spend Less: Hidden-State Probes as a Practical Recipe for Streaming Moderation of LLM Outputs
Jun 9, 2026Huizhen Shu, Xuying Li, Piao XueLLM Guardrails Safety Filtering
ModelOneAI · yunshanai
Who Earns the Safety? Intervention-Aware Quantum Predictive Control with Safety Attribution
Jun 8, 2026Yifan WangSafety-Critical Control Control Barrier Functions
Department of Mechanical Engineering, McGill University Montréal, QC H3A 2T7, Canada
Your Model Already Knows: Attention-Guided Safety Filter for Vision-Language-Action Models
Jun 8, 2026Seongbin Park, Fan Zhang, Baharan Mirzasoleiman +2Robot Safety Safety Filtering
University of California Los Angeles United States
Shield-Loco: Shielding Locomotion Policies with Predictive Safety Filtering
Jun 5, 2026Aditya Shirwatkar, Sebastian Sanokowski, Shishir Kolathaya +2Robot Safety Safety Filtering
Robert Bosch Center for Cyber Physical Systems, Indian Institute of Science, Bangalore, India · Munich Institute of Robotics and Machine Intelligence (MIRMI), Technical University of Munich, Munich, Germany · Department of Computer Science & Automation, Indian Institute of Science, Bangalore, India +2
Epistemic Injustice in Language Models: An Audit of Pretraining Filters and Guardrails
Jun 4, 2026Marco Antonio Stranisci, A Pranav, Rossana Damiano +2LLM Guardrails Training Data Curation
University of Turin, Italy · Trustworthy AI Lab, University of Hamburg, Germany · IT University of Copenhagen, Denmark
Membrane: A Self-Evolving Contrastive Safety Memory for LLM Agent Defense
Jun 4, 2026Minseok Choi, Seungbin Yang, Dongjin Kim +5LLM Guardrails Safety Filtering
♡KAIST AI · ♣Financial Tech Lab, KakaoBank Corp
Which Defense Closes Which Threat? Attributing OWASP-LLM-Top-10 Coverage and Its Brittleness Under Paraphrasing
Jun 1, 2026Alexandre Cristovão MaioranoLLM Red Teaming AI Agent Security Benchmarks
Lumytics
Permissive Safety Through Trusted Inference: Verifiable Belief-Space Neural Safety Filters for Assured Interactive Robotics
Jun 1, 2026Haimin HuRobotics Safety Filtering
Department of Computer Science, Johns Hopkins University, USA
SentGuard: Sentence-Level Streaming Guardrails for Large Language Models
Jun 1, 2026Jiaqi Yu, Xin Wang, Yixu Wang +4Language Model Safety Evaluation LLM Guardrails
1Fudan University · 2Shanghai AI Laboratory
DataShield: Safety-degrading Data Filtering for LLM Benign Instruction Fine-Tuning
May 29, 2026Junbo Zhang, Qianli Zhou, Xinyang Deng +3LLM Fine-Tuning Instruction-Tuning Data Selection
Triaging Threats to Specialized Guardrails
May 29, 2026Wenjie Jacky Mo, Xiaofei Wen, Rui Cai +6LLM Safety Benchmarks LLM Guardrails
University of California, Davis
Opir: Efficient Multi-Task Safety Classification for Toxicity, Jailbreaks, Hate Speech, and Harmful Content
May 28, 2026Ihor Stepanov, Aleksandr SmechovLanguage Model Safety Evaluation LLM Guardrails
1Knowledgator, Kyiv, Ukraine · 2Wordcab, New York, USA
FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents
May 26, 2026Haoxuan Jia, Yang Liu, Bin Chong +10AI Agent Reliability LLM Guardrails
Nanyang Technological University · Tsinghua University · Peking University +8
Robust Koopman Control Barrier Filters for Safe Actor-Critic Reinforcement Learning
May 26, 2026Dhruv S. Kushwaha, Zoleikha A. BironKoopman Operator Learning Control Barrier Functions
Reflect-Guard: Enhancing LLM Safeguards against Adversarial Prompts via Logical Self-Reflection
May 24, 2026Lixing Lin, Juli You, Yue Li +4Language Model Safety Evaluation LLM Guardrails
yaleYale University, New Haven, CT, USA · indIndependent Researcher · columbiaColumbia University, New York, NY, USA +2
Safety-Critical Control for Smoothed Implicit Contact Dynamics
May 20, 2026Haegu Lee, Yitaek Kim, Christoffer SlothDynamical Systems Contact-Rich Robotic Manipulation
The Maersk Mc-Kinney Moller Institute, University of Southern Denmark, Denmark
RefusalBench: Why Refusal Rate Misranks Frontier LLMs on Biological Research Prompts
May 20, 2026Lukas Weidener, Marko Brkić, Mihailo Jovanović +2LLM Safety Benchmarks Language Model Safety Evaluation
Applied Scientific Intelligence, Inc.
Exploring and Developing a Pre-Model Safeguard with Draft Models
May 19, 2026Hongyu Cai, Arjun Arunasalam, Yiming Liang +2LLM Guardrails Safety Filtering
Purdue University West Lafayette, Indiana, USA · Florida International University Miami, Florida, USA
Adversarial Stress Testing of SPARK Humanoid Safety Filters
May 18, 2026Saurav Ghosh, Abdou Sow, Luke ZhangAdversarial Robustness Robotics
aDepartment of Computer Science and Engineering, Washington University in St. Louis, Missouri, United States
ESLD (External Surrogate Latent Defense): A Latent-Space Architecture for Faster, Stronger Prompt-Injection Defense
May 18, 2026Yash NarendraAI Agent Security Safety Filtering
Microsoft
SafeLens: Deliberate and Efficient Video Guardrails with Fast-and-Slow Screening
May 17, 2026Shahriar Kabir Nahin, Hadi Askari, Muhao Chen +1Adaptive Inference Efficient VLM Inference
University of South Florida · University of California, Davis
Policy Library CBF: Finite-Horizon Safety at Runtime via Parallel Rollouts
May 15, 2026Taekyung Kim, Hideki Okamoto, Bardh Hoxha +2Safety-Critical Control Control Barrier Functions
Department of Robotics, 3Department of Aerospace Engineering, University of Michigan, Ann Arbor, MI, 48109, USA · Toyota Motor North America, Research & Development, Ann Arbor, MI, 48105, USA

Jul 2, 2026, cs.LGJ/K move · Enter open · S save

kNNGuard: Turning LLM Hidden Activations into a Training-Free Configurable Guardrail

Mahmoud Abdelfattah, Hamid Nasiri, Peter Garraghan

Lancaster University · Mindgard

LLM Guardrails Safety Filtering LLM Safety

Open paper AskPDF arXiv

Large language models (LLMs) are increasingly deployed in domains requiring guardrails to detect unsafe, off-topic, or adversarial prompts. Existing guardrails predominantly rely on fine-tuning to build classifiers, which often suffer from low generalization and high inference latency. We present kNNGuard, a training-free guardrail that utilizes the activation space of an off-the-shelf LLM. Given a small bank of 50 safe and unsafe prompts, kNNGuard extracts hidden activations and performs multi-layer kNN fusing activation-space and embedding-space scores for classification. Across six domains spanning topical and security prompts, kNNGuard achieves competitive or superior F1 compared to fine-tuned state-of-the-art guardrails while running 2.7x faster than the best comparable guardrail, and 10x faster than a fine-tuned safety classifier without gradient updates or fine-tuning. Domain adaptation requires only updating the labeled bank, which can be constructed in under 10 seconds and several orders of magnitude faster than established guardrails. We also analyze the impact of system prompts, layer selection, and integration into production LLM pipelines as a configurable, low-latency guardrail.

Momentum

Related topics

Latest papers 111