LLM Safety

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

36 papers in the last four weeks, up 200% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 272

All topics

SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing
Oct 7, 2026Hui Zhang, Yachao Yuan, Jiayun Wang +3LLM Security LLM Safety
North China Electric Power University · Southeast University · Soochow University +1
How Fragile Is On-Device Language Model Safety? Localizing Safety-Critical Parameters for Sparse Fault Analysis
Oct 6, 2026Muhammad Zeeshan Karamat, Christiana Chamon GarciaLLM Safety Language Model Safety Evaluation
Virginia Tech
AdaGuard: Enhancing Safety and Policy Compliance with Reasoning-Enabled LLM-As-A-Judge Guardrails
Oct 6, 2026Melissa Kazemi Rad, Sihui Dai, Isha Slavin +6LLM Guardrails LLM Safety
Capital One
Reading, Not Manipulating: Leveraging Router Logits for Multimodal Safety in MoE Vision-Language Models
Oct 6, 2026Ziyuan Yang, Wenxuan Ding, Shangbin Feng +1Vision-Language Models Safety Filtering
University of Washington · New York University
Safeguarding LLMs via Model-Agnostic Latent Safety Signals from Dark Knowledge
Oct 5, 2026Wonjun Lee, Kyungsik Yang, Gaeun Ji +5Adversarial Attacks on LLMs LLM Security
UNC Chapel Hill · Kyung Hee University · AIM Intelligence +1
Harmful SFT Leaves a Continuous Trace in LLM Checkpoint Updates
Oct 5, 2026Ziqun Bao, Xinyu Zhang, Yuchen Shao +1LLM Auditing LLM Safety
SEI, East China Normal University
Jailbreaking Open-Weight LLMs via Random Embedding Perturbations
Oct 5, 2026Abhinav Sudhakar Dubey, Scott Sirri, Vaggos Chatziafratis +1Adversarial Attacks Adversarial Attacks on LLMs
University of California Santa Cruz
TrustMI: Causally controlling how assistants trust their users
Oct 5, 2026Théo Lasnier, Romain Froger, Maxence Lasbordes +1LLM Safety Trust in AI
Inria Paris · Sorbonne Université · Meta SuperIntelligence Labs +1
Don't Judge an LLM Only by Its Activations: Discovering Suppressed Safety Features via Counterfactual Activation Potential
Oct 4, 2026Swadesh Swain, Sanghamitra DuttaAdversarial Attacks on LLMs Mechanistic Interpretability
University of Maryland, College Park
Reflections and Fragments: Securing LLMs Against Sequential Mosaic Attacks
Oct 4, 2026Emanuele La Malfa, Saar Cohen, Gabriele La Malfa +4Adversarial Attacks on LLMs LLM Safety
Department of Computer Science, University of Oxford · Institute for Decentralized AI (IDAI) · Department of Informatics, King’s College London +2
Safe Context Switching for Agents in the Wild: Mitigating Subspace Interference via Orthogonal Adaptation
Oct 4, 2026Akash Das, Ishan RoyLLM Safety Alignment LLM Safety
Fidelity Investments
FORGE: Verification-Gated Behavioral Repair for Generative Language Models
Oct 4, 2026Hsin-Ling Hsu, Min-Yu Chen, Nai-Chia Chen +3Social Bias in Language Models LLM Safety
Department of Management Information Systems National Chengchi University Taipei, Taiwan
MOMAT: Mixture of Multiple Atlases for Low-Power Jailbreak Defense of Quantized LLMs
Oct 1, 2026Boyang Li, Bingyu Shen, Weihao Hong +6Retrieval-Augmented Generation LLM Safety
Department of Computer Science, Kean University, 1000 Morris Ave, Union, 07083, NJ, USA · Department of Computer Science and Engineering, University of Notre Dame, 257 Fitzpatrick Hall of Engineering, Notre Dame, 46556, IN, USA · McGill University, 845 Sherbrooke Street, Montréal, H3A 0G4, Québec, Canada +2
Preemptive LLM Unlearning against Forbidden Capability Acquisition via Gradient Sealing
Sep 30, 2026Kemou Li, Qizhou Wang, Yue Wang +6LLM Security LLM Safety
State Key Laboratory of Internet of Things for Smart City, University of Macau · RIKEN Center for Advanced Intelligence Project · The University of Melbourne +5
MASCRDM: Multi-Agent System for Compliance Risk Detection and Mitigation in Training Process of Large Language Models
Sep 30, 2026Yan Zhang, Chuming Wei, Ruien Li +3LLM Safety Language Model Bias Evaluation
Department of Computer Science and Technology, Tsinghua University · Academy of Artificial Intelligence and Advanced Technology, Xi’an Jiaotong-Liverpool University · Department of Computer Sciences, University of Wisconsin–Madison +1
ThinkingGuard: Decoding Implicit Hazards via Step-by-Step Risk Attribution in Multimodal Large Language Models
Sep 29, 2026Ruochen Zhang, Yao Huang, Yitong Sun +5Multimodal Reasoning LLM Safety
Beihang University Beijing, China · Tsinghua University Beijing, China · Institute of Artificial Intelligence Beihang University Beijing, China State Key Laboratory of AI Safety
Guard Models Are Overconfident Where Base Models Are Uncertain
Sep 29, 2026Jonghyun Hong, MinJae Jung, Minwoo KimLanguage Model Safety Evaluation Adversarial Attacks on LLMs
DATUMO INC.
The Safety Operator: Modulating the Expression of Safety Instructions via Spectral Optimization
Sep 29, 2026Benoit Dherin, Michael Munn, Xavier Gonzalvo +14LLM Safety Alignment LLM Safety
Google Research · Cambridge University · Google DeepMind +2
Right Words, Wrong Moment: A Clinician-Grounded Analysis of Distress in 19,930 Conversations between Young People and ChatGPT
Sep 28, 2026Marx Wang, Ella Zhang, Cameron Tan +11Emotional Support Conversation Human-AI Interaction
University of Washington, United States · University of Oxford, United Kingdom · Stanford University, United States +2
Applying Language Models in Clinical Medicine: Recent Trends and Perspectives
Sep 28, 2026Erik AertsLLM Reliability LLM Safety
When Do Model Internals Help? Exploring the Role of Representation Engineering in LLM Safety
Sep 28, 2026Tianyi Guan, Jianhui Chen, Liangming PanLanguage Model Safety Evaluation LLM Interpretability
State Key Laboratory of Multimedia Information Processing, Peking University · School of Computer Science, Peking University
How to Tame a Multi-Headed Hydra? Adaptive Multi-Category Safety Steering for Large Language Models
Sep 28, 2026Chenxi Wang, Ruiyang Huang, Li Huang +1Language Model Steering LLM Safety
Southeast University, Nanjing, China · Peking University, Beijing, China · Chongqing University, Chongqing, China
AdaGuard: An Adaptive Guard Model with User-defined Policies
Sep 28, 2026Yunhao Feng, Yifan Ding, Yuxiang Xie +4LLM Guardrails AI Agent Safety
National University of Defense Technology · Fudan University
Quantifying Behavioral Tails in Black-Box Language Models
Sep 27, 2026Elsayed Eshra, Ali Al-Lawati, Dongwon Lee +1Language Model Safety Evaluation LLM Safety
The Pennsylvania State University University Park, PA 16802, USA
Beyond Average Safety: Chance-Constrained LLM Fine-tuning
Sep 24, 2026Taha Entesari, Mahyar FazlyabStochastic Optimization Chance-Constrained Optimization
Department of Electrical and Computer Engineering Johns Hopkins University
Rare Event Estimation via Iterative Unalignment
Sep 21, 2026Hanming Yang, Daksh Mittal, Jing Dong +1Importance Sampling LLM Safety
Decision, Risk, and Operations Division, Columbia Business School
Measuring the Assistant's Harmlessness Preferences on the User Turn
Sep 20, 2026Jord NguyenLanguage Model Safety Evaluation LLM Safety
Harm Laundering in GPT Models: Evidence That Gender Discrimination Is Transformed Rather Than Reduced Across Safety-Trained Generations
Sep 17, 2026Sarah Wyer, Sue Black, Noura Al MoubayedGender Bias in Language Models Language Model Safety Evaluation
Durham University
Local Sparsity Enables Unsupervised LLM Safety Detection
Sep 17, 2026Xin Chen, Gil Kur, Alexander Shevchenko +1Sparse Autoencoders LLM Safety
ETH Zürich
Safety Beyond the Interface: Detecting Harm via Latent States in Large Language Models
Sep 16, 2026Alizishaan Khatri, Chiquita Prabhu, Omkar NeogiSafety Filtering LLM Safety
Wrynx Inc. · Independent Researcher

Oct 7, 2026, cs.CRJ/K move · Enter open · S save

SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing

Hui Zhang, Yachao Yuan, Jiayun Wang, Yuanzhuo Li+2

North China Electric Power University · Southeast University · Soochow University+1

LLM Security LLM Safety LLM Safety Alignment

Open paper AskPDF arXiv

Fine-tuning-as-a-service enables users to adapt aligned large language models (LLMs) to specialized tasks, but malicious fine-tuning can erode refusal behavior while preserving task performance on legitimate inputs. We revisit recent layer-wise safety diagnostics and find that safety sensitivity is signed: scaling different layers can strengthen refusal, weaken it, or have little effect. Motivated by this observation, we propose SLDR, a post-fine-tuning defense based on Selective Layers Recovery and Dynamic Routing. SLDR trains a LoRA recovery adapter only on the layers with the maximum and minimum sensitivity scores in the signed spectrum, and uses representation-based dynamic routing inference to activate the adapter only for malicious queries. Across four model architectures, five downstream tasks, and four harmful benchmarks, SLDR substantially reduces harmful outputs while preserving downstream utility. On Llama3.1/SST2, SLDR reduces the average harmful score from 11.54 to 0.08 while maintaining downstream accuracy, and the harmful score remains near zero under poisoning ratios up to 0.9. The code is available at https://github.com/Stardust457/SLDR.

Momentum

Related topics

Latest papers 272