Safety Alignment

Follow this field Ask about this field

Recent momentum

emerging

0 papers in the last 28 days · 0.0% of indexed attention

Twelve weeks of publication activity for this topic as it is defined today.

Weekly history

Recent digests

What was published in this field, kept on the site without email delivery.

Period ending 2026-09-21

16 new papers

A weekly snapshot of new work published in Safety Alignment.

Period ending 2026-09-14

14 new papers

A weekly snapshot of new work published in Safety Alignment.

Period ending 2026-09-07

25 new papers

A weekly snapshot of new work published in Safety Alignment.

Inside this field

Focused directions

Focused topic

Safety

Within Safety Alignment

291 papersExplore

Focused topic

Safety Alignment

Within Safety Alignment

94 papersExplore

Focused topic

Safety Evaluation

Within Safety Alignment

83 papersExplore

Focused topic

Safeguard

Within Safety Alignment

61 papersExplore

464 papers

Latest in Safety Alignment

Browse all topics →

LLM Safety From Within: Detecting Harmful Content with Internal Representations
Apr 20, 2026Difan Jiao, Yilun Liu, Ye Yuan +4Large Language Model Safety Harmful Content
Using large language models for embodied planning introduces systematic safety risks
Apr 20, 2026Tao Zhang, Kaixian Qu, Zhibin Li +4Large Language Model Planning Classical Planning
Owner-Harm: A Missing Threat Model for AI Agent Safety
Apr 20, 2026Dongcheng Zhang, Yiqing JiangArtificial Intelligence Safety Threat Models
MHSafeEval: Role-Aware Interaction-Level Evaluation of Mental Health Safety in Large Language Models
Apr 20, 2026Suhyun Lee, Palakorn Achananuparp, Neemesh Yadav +2Large Language Model Safety Large Language Model Evaluation
SafeAnchor: Preventing Cumulative Safety Erosion in Continual Domain Adaptation of Large Language Models
Apr 20, 2026Dongxin Guo, Jikun Wu, Siu Ming YiuSafety Alignment Large Language Model Adaptation
The Open-Weight Paradox: Why Restricting Access to AI Models May Undermine the Safety It Seeks to Protect
Apr 19, 2026Vinicius Santana GomesArtificial Intelligence Governance Governance Layer
Cat-DPO: Category-Adaptive Safety Alignment
Apr 19, 2026Tiankai Yang, Yi Nian, Xinyuan Li +3Safety Alignment Large Language Model Alignment
Continual Safety Alignment via Gradient-Based Sample Selection
Apr 19, 2026Thong Bach, Dung Nguyen, Thao Minh Le +1Safety Alignment Large Language Model Alignment
Guardrails in Logit Space: Safety Token Regularization for LLM Alignment
Apr 19, 2026Thong Bach, Truyen TranSafety Alignment Large Language Model Alignment
A Real-Time Bike-Pedestrian Safety System with Wide-Angle Perception and Evaluation Testbed for Urban Intersections
Apr 18, 2026Mehmet Kerem TurkcanCollision Prediction Safety-Critical Scenarios
On Safety Risks in Experience-Driven Self-Evolving Agents
Apr 18, 2026Weixiang Zhao, Yichen Zhang, Yingshuo Wang +8Self-Evolving Agents Safety
Vision-Based Human Awareness Estimation for Enhanced Safety and Efficiency of AMRs in Industrial Warehouses
Apr 18, 2026Maximilian Haug, Christian Stippel, Lukas Pscherer +5Human-Robot Collaboration Robotic Perception
When Choices Become Risks: Safety Failures of Large Language Models under Multiple-Choice Constraints
Apr 18, 2026Yuheng Chen, Zhiyu Wu, Bowen Cheng +1Large Language Model Safety Safety Alignment
SafeDream: Safety World Model for Proactive Early Jailbreak Detection
Apr 18, 2026Bo Yan, Weikai Lin, Yada Zhu +1Jailbreak Detection Jailbreak Attacks
Benign Fine-Tuning Breaks Safety Alignment in Audio LLMs
Apr 17, 2026Jaechul Roh, Amir HoumansadrLarge Audio Language Models Audio Flow
SafeLM: Unified Privacy-Aware Optimization for Trustworthy Federated Large Language Models
Apr 17, 2026Noor Islam S. Mohammad, Uluğ BayazıtLarge Language Model Safety Safeguard
MCBench: A Multicontext Safety Assessment Benchmark for Omni Large Language Models
Apr 17, 2026Manh Luong, Tamas Abraham, Junae Kim +6Large Language Model Safety Safety Evaluation
Pruning Unsafe Tickets: A Resource-Efficient Framework for Safer and More Robust LLMs
Apr 17, 2026Wai Man Si, Mingjie Li, Michael Backes +1Large Language Model Safety Large Language Model Alignment
Agentic Microphysics: A Manifesto for Generative AI Safety
Apr 16, 2026Federico Pierucci, Matteo Prandi, Marcantonio Bracale Syrnikov +2Artificial Intelligence Safety Agentic Systems
Benchmarks for Trajectory Safety Evaluation and Diagnosis in OpenClaw and Codex: ATBench-Claw and ATBench-Codex
Apr 16, 2026Zhonghao Yang, Yu Li, Yanxu Zhu +6Safety Evaluation Agentic Benchmarks
HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark
Apr 15, 2026Jiacheng Wang, Jinchang Hou, Fabian Wang +3Safety Evaluation Trajectory Prediction
Safety Training Modulates Harmful Misalignment Under On-Policy RL, But Direction Depends on Environment Design
Apr 14, 2026Leon Eshuijs, Shihan Wang, Antske FokkensSafety Alignment Safety
SafeMath: Safe Solutions for Unsafe Math Word Problems
Mar 26, 2026Sagnik Basu, Subhrajit Mitra, Aman Juneja +3Large Language Model Safety Math Problems
SafeFlow: Real-Time Text-Driven Humanoid Whole-Body Control via Physics-Guided Rectified Flow and Selective Safety Gating
Mar 25, 2026Hanbyel Cho, Sang-Hun Kim, Jeonguk Kang +1Long-Tail Humanoid Whole-Body Control Robot Trajectories
Algorithms for Deciding the Safety of States in Fully Observable Non-deterministic Problems: Technical Report
Mar 16, 2026Johannes Schmalz, Chaahat JainSafety Constraints Worst-Case
Real-Time Driver Safety Scoring Through Inverse Crash Probability Modeling
Mar 16, 2026Joyjit Roy, Samaresh Kumar Singh, Sushanta Das +1Accidents Crashes
PC-Diffuser: Path-Consistent Capsule CBF Safety Filtering for Diffusion-Based Trajectory Planner
Mar 11, 2026Eugene Ku, Yiwei LyuDiffusion Planning Autonomous Driving
Mitigating the Safety-utility Trade-off in LLM Alignment via Adaptive Safe Context Learning
Feb 14, 2026Yanbo Wang, Minzheng Wang, Jian Liang +3Safety Alignment Large Language Model Alignment
Multimodal Benchmark for Safety Assessment in Industrial Inspection Scenarios
Jan 29, 2026Zeyi Liu, Shuang Liu, Jihai Min +7Inspection Multi-Modal Sensing
Preserving Fairness and Safety in Quantized LLMs Through Critical Weight Protection
Jan 17, 2026Muhammad Alif Al Hakim, Alfan Farizki Wicaksono, Fajri KotoLarge Language Model Quantization Large Language Model Safety
Towards Multi-modal Multi-turn Safety: From Agentic Interaction to Strategic Alignment
Jan 8, 2026Han Zhu, Jiale Chen, Chengkun Cai +8Safety Alignment Multi-Modal Large Language Models
Safety Is Not Universal: The Selective Safety Trap in LLM Alignment
Jan 7, 2026Iago Alves Brito, Walcy Santos Rezende Rios, Julia Soares Dollis +2Large Language Model Safety Large Language Model Alignment
Check Yourself Before You Wreck Yourself: Selectively Quitting Improves LLM Agent Safety
Oct 18, 2025Vamshi Krishna Bonagiri, Ponnurangam Kumaragurum, Khanh Nguyen +1Large Language Model Agents Large Language Model Safety
Attention Misses Visual Risk: Risk-Adaptive Steering for Multimodal Safety Alignment
Oct 15, 2025Jonghyun Park, Minhyuk Seo, Chaewon Yeo +1Inference-Time Defense Safety Alignment
Turning the Spell Around: Lightweight Alignment Amplification via Rank-One Safety Injection
Aug 28, 2025Harethah Abu Shairah, Hasan Abed Al Kader Hammoud, George Turkiyyah +1Safety Alignment Large Language Model Alignment
Integrated Noise and Safety Management in UAM via A Unified Reinforcement Learning Framework
Aug 22, 2025Surya Murthy, Zhenyu Gao, John-Paul Clarke +1Multi-Uav Systems Air Traffic Control
Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Models
Jun 9, 2025Mickel Liu, Liwei Jiang, Yancheng Liang +4Large Language Model Safety Safety Alignment
Captivity-Escape Games as a Means for Safety in Online Motion Generation
Jun 2, 2025Christopher Bohn, Manuel Hess, Sören HohmannSafety Constraints Escape
SHIELD: Safety on Humanoids via CBFs In Expectation on Learned Dynamics
May 16, 2025Lizhi Yang, Blake Werner, Ryan K. Cosner +3Control Barrier Functions Scalable Robot Learning
Learning-based Adaptive Safety-Critical Control With Evolving Unsafe Regions
Mar 19, 2025Songqiao Hu, Zidong Wang, Zeyi Liu +2Control Barrier Functions Safety-Critical Scenarios
Real-Time Safety Evaluation of Human Arm Operations Using a Wrist-Mounted IMU with PSM System
Feb 13, 2025Musab Zubair Inamdar, Alhusain Al Hadrami, Seyed Amir TafrishiHuman-Robot Collaboration Robotic Arms
Who Judges the Judges? A Chinese Safety QA Benchmark for Evaluating LLM Responses and Safety Judges
Date pendingRui Yang, Shuang Huang, Junhua Liu +7Large Language Model Safety Safety Evaluation
Layered Safety: Enhancing Autonomous Collision Avoidance via Multistage CBF Safety Filters
Date pendingErina Yamaguchi, Ryan M. Bena, Gilbert Bahati +1Collision Avoidance Safety Filters
SAFER-Activities: A Dataset for Smart Assessment of Fall Events and Routine Activities
Date pendingDiwas Lamsal, Pramod Wickramatilake, Jednipat Moonrinta +2Robust Fall Detection Human Activity Recognition

Apr 20, 2026 · cs.AIJ/K move · Enter open · S save

LLM Safety From Within: Detecting Harmful Content with Internal Representations

Difan Jiao, Yilun Liu, Ye Yuan, Zhenwei Tang+3

University of Toronto · LMU Munich · McGill University

Guard models are widely used to detect harmful content in user prompts and LLM responses. However, state-of-the-art guard models rely solely on terminal-layer representations and overlook the rich safety-relevant features distributed across internal layers. We present SIREN, a lightweight guard model that harnesses these internal features. By identifying safety neurons via linear probing and combining them through an adaptive layer-weighted strategy, SIREN builds a harmfulness detector from LLM internals without modifying the underlying model. Our comprehensive evaluation shows that SIREN substantially outperforms state-of-the-art open-source guard models across multiple benchmarks while using 250 times fewer trainable parameters. Moreover, SIREN exhibits superior generalization to unseen benchmarks, naturally enables real-time streaming detection, and significantly improves inference efficiency compared to generative guard models. Overall, our results highlight LLM internal states as a promising foundation for practical, high-performance harmfulness detection.

Large Language Model Safety Harmful Content Safety

Open paper Ask about it PDF