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 →

PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models
Jun 24, 2026Chang Wu, Junfeng Fang, Houcheng Jiang +5Safety Alignment Large Language Model Safety
Do Thinking Tokens Help with Safety?
Jun 23, 2026Narutatsu Ri, Abhishek Panigrahi, Sanjeev AroraLarge Reasoning Models Thinking
AutoSpec: Safety Rule Evolution for LLM Agents via Inductive Logic Programming
Jun 23, 2026Pingchuan Ma, Zhaoyu Wang, Zimo Ji +5Safety Large Language Model Agents
LIBERO-Safety: A Comprehensive Benchmark for Physical and Semantic Safety in Vision-Language-Action Models
Jun 22, 2026Rongxu Cui, Zongzheng Zhang, Jingrui Pang +11Vision-Language-Action Models Safety Evaluation
Evaluation Awareness Is Not One Capability: Evidence from Open Language Models
Jun 22, 2026Nilesh Nayan, Aishwarya Sampath Kumar, Rishiraj Girmal +5Safety Evaluation Large Language Model Safety
IndicGuard: A Multilingual Safety Guard Model and Dataset for Indic Languages
Jun 22, 2026Parth Bramhecha, Smit Deshmukh, Sairaj Bodhale +2Multilingual Safety Large Language Model Safety
Verifiable Foundation Models for Robot Safety
Jun 22, 2026Davide Corsi, Kyungmin Kim, Roy FoxScalable Robot Learning Robot Systems
Safety-Aware Evaluation of LLM-Generated Driver Intervention Messages through Multi-Task Risk Fusion
Jun 21, 2026Keito InoshitaSafety Evaluation Llm-Driven
AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent
Jun 21, 2026Weidi Luo, Qiming Zhang, Yihao Quan +5Large Language Model Safety Safety
Imagine to Ensure Safety in Hierarchical Reinforcement Learning
Jun 21, 2026Gregory Gorbov, Artem Latyshev, Aleksandr I. PanovHierarchical Reinforcement Learning Safety Constraints
Enhancing Road Safety: An IoT-Based Accident Detection and Prevention Mechanism
Jun 21, 2026Prabhu Pugalenthi, Pramod Krishnaa DhanbalanDriving Early Warning
EmbodiedUS-FS: Fast Slow Intelligence for Ultrasound Robotics
Jun 21, 2026Fangzhuo Zhang, Xinyu Wang, Xiao Yang +1Ultrasound Robotics
Any-Body Guard: Universal Safeguarding for Manipulation Policies via Action Masking
Jun 21, 2026Alex Beaudin, Hanna Krasowski, Kartik Nagpal +3Collision Avoidance Safety Constraints
Safe responses matter: Output-aware safety guardrail mitigate over-refusal in MLLMs
Jun 20, 2026Jiayi Li, Kun ZhanLarge Language Model Safety Safety
CNnotator: LLM-Guided Memory Safety Annotation Synthesis
Jun 20, 2026Twain Byrnes, Mike DoddsProgram Analysis Large Language Model Safety
NRT-Bench: Benchmarking Multi-Turn Red-Teaming of LLM Operator Agents in Safety-Critical Control Rooms
Jun 18, 2026Hanwool Lee, Dasol Choi, Bokyeong Kim +2Red-Teaming Safety Evaluation
SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling
Jun 18, 2026Haotian Xu, Zeyang Zhang, Linbao Li +3Inference-Time Defense Speculative Decoding
Beyond Safe Data: Pretraining-Stage Alignment with Regular Safety Reflection
Jun 17, 2026Jinhan Li, Kexian Tang, Yihan Xu +2Safety Alignment Large Language Model Alignment
ROBOSHACKLES: A Safety Dataset for Human-Injury Prevention in Embodied Foundation Models
Jun 17, 2026Zhuowen Yin, Chongyang Liu, Wenzhang Yang +2Robotic Datasets Safety-Critical Scenarios
Towards Understanding and Measuring COGNITIVE ATROPHY in LLM Behaviour
Jun 16, 2026Abeer Badawi, Moyosoreoluwa Olatosi, Negin Baghbanzadeh +5Safety Evaluation Mild Cognitive Impairment
Greed Is Learned: Visible Incentives as Reward-Hacking Triggers
Jun 15, 2026Tong Che, Rui WuIncentives Shared Rewards
Is Your Trajectory Displacement Safe in Long-tail?
Jun 15, 2026Qiao Sun, Weicheng Zheng, Yixin Huang +1Autonomous Driving Safety Evaluation
Rapid Poison: Practical Poisoning Attacks Against the Rapid Response Framework
Jun 15, 2026David Huang, Jaewon Chang, Avidan Shah +2Data Poisoning Backdoor Attacks
CHILLGuard: Towards Fine-Grained Chinese LLM Safety Guardrail with Scalable Data Construction and Model-aware Preference Alignment
Jun 13, 2026Wenbo Yu, Bohua Wang, Hao Fang +10Large Language Model Safety Safety
OSGuard: A Benchmark for Safety in Computer-Use Agents
Jun 13, 2026Mina Mohammadmirzaei, Jeffrey FlaniganSafety Evaluation Computer-Use Agents
Safe Reinforcement Learning of Autonomous Highway Driving: A Unified Framework for Safety and Efficiency
Jun 12, 2026Chufei Yan, Zhihao Cui, Yiyan Lv +3Autonomous Driving Deep Reinforcement Learning
ForceForget: Reinforcement Concept Removal for Enhancing Safety in Text-to-Image Models
Jun 12, 2026Dong Han, Yong LiText-To-Image Adversarial Prompting
I'm Sorry Driver, I'm Afraid I Can't Do That: Appraising the Safety of LLMs within Automotive Contexts
Jun 12, 2026Shaun Feakins, Ibrahim Habli, Kim Littler +1Large Language Model Safety Llm-Driven
When and How Severely: Scenario-Specific Safety Envelopes for Driving VLAs
Jun 12, 2026Abhinaw Priyadershi, Jelena FrtunikjDriving Safety-Critical Scenarios
Safeguard-Conditioned Uplift: Measuring Utility-Risk Frontiers for Dual-Use Biology Assistants
Jun 12, 2026Dipesh Tharu MahatoSafety Evaluation Pareto Frontier
Designing Safety-Constrained LLM Systems for Public Health Information Access
Jun 11, 2026Ben Torkian, Jun ZhouLarge Language Model Safety Safety
SafeLLM: Extraction as a Hallucination-Resistant Alternative to Rewriting in Safety-Critical Settings
Jun 11, 2026Julia Ive, Felix Jozsa, Evridiki Georgaki +6Large Language Model Safety Large Language Model Hallucination
The Containment Gap: How Deployed Agentic AI Frameworks Fail Public-Facing Safety Requirements
Jun 11, 2026Md Jafrin Hossain, Mohammad Arif Hossain, Weiqi Liu +1Large Language Model Safety Modular Agentic Framework
Online Shift Detection and Conformal Adaptation for Deployed Safety Classifiers
Jun 10, 2026Jun Wen LeongDistribution Shifts Attack Detection
SAFER-Nav: Enhancing Safety for Visual Robot Navigation via Segmentation-Aware Fine-Tuning
Jun 10, 2026Geonyeong Ko, Giung Lee, Changjoo NamSocial Robot Navigation Obstacle Avoidance
Quality Is Not a Safety Proxy Under Quantization
Jun 8, 2026Sahil KadadekarQuantization Safety Evaluation
Who Earns the Safety? Intervention-Aware Quantum Predictive Control with Safety Attribution
Jun 8, 2026Yifan WangSafety Filters Model Predictive Control
Your Model Already Knows: Attention-Guided Safety Filter for Vision-Language-Action Models
Jun 8, 2026Seongbin Park, Fan Zhang, Baharan Mirzasoleiman +2Vision-Language-Action Models Large Vision Language Models
Distilling Safe LLM Systems via Soft Prompts for On Device Settings
Jun 8, 2026Motasem Alfarra, Cristina Pinneri, Dana Kianfar +2Large Language Model Safety Safety Alignment
Reliable to Expressive: A Curriculum for Rubric-Following Safety Judges
Jun 8, 2026Yongtaek Lim, Hyeji Choi, Minwoo KimSafety Evaluation Rubric Framework
Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs
Jun 8, 2026Yanyan Luo, Xue Han, Ruiqiao Bai +10Large Language Model Safety Personalization
SafeRun: Enabling Determinism in LLM Planning for Running
Jun 8, 2026Meilin Chen, Zepeng Zhai, Jiaxuan Zhao +1Large Language Model Planning Classical Planning
CARE: A Conformal Safety Layer for Medical Summarization
Jun 8, 2026Suhana Bedi, Bridget Lin, Anson Y. Zhou +5Summarization Large Language Model Safety
Safe, Fluent and Acceptable Motion Generation and Execution for Human--Robot Interaction in Manufacturing Environments
Jun 7, 2026Thibaut Lopez, Olivier Aycard, Pierre-Brice Wieber +2Human-Robot Interaction Human-Centered Framework
SSAFE: Simple and Strong AI-Generated Image Detection via Frozen Vision Encoders
Jun 7, 2026Seunghyun Lee, Byoungkwon Kim, Jaehyun Nam +2Ai-Generated Image Detection Explainable Deepfake Detection
ForesightSafety-SAGE:A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents
Jun 7, 2026Lu Jia, Haibo Tong, Feifei Zhao +5Large Language Model Safety Large Language Model Agents
Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and Models
Jun 7, 2026Arya Shah, Himanshu Beniwal, Mayank Singh +1Multilingual Safety Cross-Lingual Alignment
PACT: Self-Evolving Physical Safety Alignment for Diffusion Policies in Embodied Manipulation
Jun 7, 2026Lingxuan Wu, Zijian Zhu, Lizhong Wang +5Diffusion Policies Safety Constraints
SciTrace: Trajectory-Aware Safety Reasoning for Scientific Discovery Agents
Jun 6, 2026Tanush Swaminathan, Runmin Jiang, Letian Zhang +1Scientific Agents Safety
When Behavioral Safety Evaluation Fails: A Representation-Level Perspective
Jun 6, 2026Enyi Jiang, Anders Gjølbye, Yibo Jacky Zhang +1Large Language Model Safety Large Language Model Evaluation
Enhancing AI Interpretability and Safety through Localised Architectures
Jun 6, 2026Ian Seet, Jonas Bozenhard, Simon OstermannModel Interpretability Neural Architectures
Safety is Contextual, LLM-Judges Are Not: Navigating the Rigid Priors of Evaluators
Jun 5, 2026Anissa Alloula, Federico Licini, Ava Batchkala +1Llm-As-A-Judge Judges
The Cold-Start Safety Gap in LLM Agents
Jun 5, 2026Chung-En Sun, Linbo Liu, Tsui-Wei WengLarge Language Model Agents Large Language Model Safety
Shield-Loco: Shielding Locomotion Policies with Predictive Safety Filtering
Jun 5, 2026Aditya Shirwatkar, Sebastian Sanokowski, Shishir Kolathaya +2Safety Filters Perceptive Locomotion Policies
Extending Responsibility-Sensitive Safety for the Assessment of Offloaded Autonomous Driving Services
Jun 5, 2026Robin Dehler, Aryan Thakur, Michael BuchholzAutonomous Driving Offloading
DOG-DPO:Dynamic Optimization in Geometry for Safety Alignment
Jun 4, 2026Yi Nian, Tiankai Yang, Yudi Zhang +7Safety Alignment Large Language Model Alignment
SpeechJBB: Probing Safety Alignment and Comprehension in Large Audio Language Models under Code-Switched Speech
Jun 4, 2026Virginia Ceccatelli, Yejin Jeon, David Ifeoluwa AdelaniLarge Language Model Jailbreaks Large Audio Language Models
Consistency Training Along the Transformer Stack
Jun 4, 2026Sukrati Gautam, Neil Shah, Arav Dhoot +7Safety Alignment Misalignment
Safe Embodied AI for Long-horizon Tasks: A Cross-layer Analysis of Robotic Manipulation
Jun 4, 2026Dabin Kim, Daemin Park, Sangyub Lee +4Generalizable Robotic Manipulations Long-Horizon Planning
Safety Paradox: How Enhanced Safety Awareness Leaves LLMs Vulnerable to Posterior Attack
Jun 4, 2026Long P. Hoang, Hai V. Le, Shaoyang Xu +2Large Language Model Safety Safety Alignment

Jun 24, 2026 · cs.CLJ/K move · Enter open · S save

PolicyAlign: Direct Policy-Based Safety Alignment for Large Language Models

Chang Wu, Junfeng Fang, Houcheng Jiang, Kai Tang+4

Qwen Large Model Application Team, Alibaba · National University of Singapore · Zhongguancun Academy

Safety alignment of large language models (LLMs) typically depends on high-quality supervision data, such as safe demonstrations or preference pairs. However, in real-world deployment, emerging safety requirements are often specified as natural-language policies, while corresponding supervision data may be costly, delayed, or unavailable. This creates a mismatch between rapidly evolving safety policies and conventional data-driven alignment methods. To address this, we propose PolicyAlign, a simple yet effective framework for directly aligning LLMs with safety policies. Given a safety policy, PolicyAlign first synthesizes policy-violating instructions and then performs on-policy self-distillation to internalize policy-guided behavior. To improve training stability and data efficiency, we further introduce Policy-Sensitive Filtering, which selects instructions where the policy induces the largest behavioral shift. Experiments across multiple models show that PolicyAlign consistently improves safety while maintaining low over-refusal and preserving general capabilities. PolicyAlign also generalizes to medical, legal, and financial safety scenarios, highlighting its potential as a scalable and maintainable approach to policy-based LLM safety alignment. The code is released at https://github.com/Qwen-Applications/PolicyAlign.

Safety Alignment Large Language Model Safety Large Language Model Alignment

Open paper Ask about it PDF