J-Anchor: Defending Against Prefill Attacks via J-Space Selective Refusal Restoration
cs.CR, cs.AI
Submitted: 2026-10-31
Updated: 2026-10-31
Terminology
Sources
- GPT-4 Technical Report
- Eliciting Latent Predictions from Transformers with the Tuned Lens
- Discovering Latent Knowledge in Language Models Without Supervision
- HARC: Coupling Harmfulness and Refusal Directions for Robust Safety Alignment
- Training Verifiers to Solve Math Word Problems
- Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models
- The Llama 3 Herd of Models
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Verbalizable Representations Form a Global Workspace in Language Models
- WildTeaming at Scale: From In-the-Wild Jailbreaks to (Adversarially) Safer Language Models
- HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
- Silent Alarm: A J-Space Protocol for Comparing Danger Recognition Across Models and Quantization Levels
- Exposing the Systematic Vulnerability of Open-Weight Models to Prefill Attacks
- Gemma 4 Technical Report
- Gemma 3 Technical Report
- Qwen3 Technical Report
- Bypassing the Safety Training of Open-Source LLMs with Priming Attacks
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- Mitigating Memorization Where It Happens
- Defending Large Language Models Against Jailbreak Attacks via In-Decoding Safety-Awareness Probing
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