Eraser: Jailbreaking Defense in Large Language Models via Unlearning Harmful Knowledge

arXiv:2404.05880 · cs.CL · Submitted 2024-04-08 · Read on arXiv

cs.CL

Submitted: 2024-04-08

Updated: 2026-09-20

Code: https://github.com/ZeroNLP/Eraser

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Alignment — 20× in this paper · explained in 1 episode
• Benchmark — 7× in this paper · explained in 2 episodes
• Accuracy — 5× in this paper · explained in 1 episode
• Machine Unlearning — 4× in this paper · explained in 2 episodes
• Ablation — 2× in this paper · explained in 1 episode
• Alignment Methods — 2× in this paper · explained in 1 episode
• Large Language Models (LLMs) — 2× in this paper · explained in 6 episodes
• Objective Function — 2× in this paper · explained in 2 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• AdamW Optimizer — 1× in this paper · explained in 1 episode
• Adaptation — 1× in this paper · explained in 1 episode
• Adversarial Prompting — 1× in this paper · explained in 1 episode
• Approximate Unlearning — 1× in this paper · explained in 1 episode
• Attack Success Rate — 1× in this paper
• Causal Reasoning — 1× in this paper · explained in 2 episodes
• Distillation — 1× in this paper · explained in 2 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Hallucination — 1× in this paper · explained in 3 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes
• Verification — 1× in this paper · explained in 2 episodes

Sources

• GPT-4 Technical Report
• PaLM 2 Technical Report
• Qwen Technical Report
• Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
• Toxicity in ChatGPT: Analyzing Persona-assigned Language Models
• Who's Harry Potter? Approximate Unlearning in LLMs
• LLM Self Defense: By Self Examination, LLMs Know They Are Being Tricked
• LoRA: Low-Rank Adaptation of Large Language Models
• Defending Against Alignment-Breaking Attacks via Robustly Aligned LLM
• A Survey of Safety and Trustworthiness of Large Language Models through the Lens of Verification and Validation
• Jailbreaking Black Box Large Language Models in Twenty Queries
• Certifying LLM Safety against Adversarial Prompting
• AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models
• Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
• SmoothLLM: Defending Large Language Models Against Jailbreaking Attacks
• Llama 2: Open Foundation and Fine-Tuned Chat Models
• Defending Large Language Models Against Jailbreaking Attacks Through Goal Prioritization
• Making Harmful Behaviors Unlearnable for Large Language Models
• Universal and Transferable Adversarial Attacks on Aligned Language Models
• Baichuan 2: Open Large-scale Language Models

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering