Safety Direction Adapter: A Training-Free Post-Hoc Safety Patch for Fine-Tuned Open-Weight LLMs
cs.CR
Submitted: 2026-11-06
Updated: 2026-11-06
Terminology
Sources
- Refusal in Language Models Is Mediated by a Single Direction
- Locking Down the Finetuned LLMs Safety
- Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
- Safety Alignment Should Be Made More Than Just a Few Tokens Deep
- Assessing the Brittleness of Safety Alignment via Pruning and Low-Rank Modifications
- Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
- How Post-Training Reshapes LLMs: A Mechanistic View on Knowledge, Truthfulness, Refusal, and Confidence
- FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning
- TrustLLM: Trustworthiness in Large Language Models
- HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
- Purple Llama CyberSecEval: A Secure Coding Benchmark for Language Models
- There Is More to Refusal in Large Language Models than a Single Direction
- LLMs Encode Harmfulness and Refusal Separately
- Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks
- Safety Layers in Aligned Large Language Models: The Key to LLM Security
- Understanding and Preserving Safety in Fine-Tuned LLMs
- Patcher: Post-Hoc Patching of Backdoored Large Language Models
- Refusal in LLMs is an Affine Function
- Anchoring Refusal Direction: Mitigating Safety Risks in Tuning via Projection Constraint
- Representation Engineering: A Top-Down Approach to AI Transparency
Related papers
- From Strategy Search to Policy Learning: MCTS-Guided Adversarial Co-Evolution for LLM Safety Alignment
- Privacy Leakage from Recurrent State in LLM Inference
- Incrementally Verifiable Oracle Aided Computation
- Privilege and confidentiality in generative AI workflows
- BadMoE: Backdooring Mixture-of-Experts LLMs via Routing-Aware Triggers and Task-Underutilized Experts
- CoVault: Secure, Scalable Analytics of Personal Data