Refusal in language models is mediated by a single direction in residual stream activations that can be erased to disable safety or added to elicit refusal.
Removing rlhf protections in gpt-4 via fine-tuning
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
citation-role summary
background 1
method 1
citation-polarity summary
polarities
background 2representative citing papers
GPT-4 LLM agents autonomously exploit 87% of tested one-day vulnerabilities when given CVE descriptions, far outperforming other models and tools.
NeWTral is a non-linear weight translation framework using MoE routing that reduces average attack success rate from 70% to 13% on unsafe domain adapters across Llama, Mistral, Qwen, and Gemma models up to 72B while retaining 90% knowledge fidelity.
REGLU guides LoRA-based unlearning via representation subspaces and orthogonal regularization to outperform prior methods on forget-retain trade-off in LLM benchmarks.
FRPO applies a max-min robust optimization over KL-bounded policy neighborhoods during RLHF to reduce catastrophic forgetting of safety and accuracy under subsequent SFT or RL fine-tuning.
StrongREJECT provides a standardized benchmark and evaluator for jailbreak attacks that aligns better with human judgments than prior methods and reveals that successful jailbreaks often reduce model capabilities.
Dominant control-based AI alignment falls short for potential AGI subjects; a parenting model drawing on Turing's child machines should foster gradual autonomy and cooperative coexistence.
A survey that creates taxonomies for jailbreak attacks and defenses on LLMs, subdivides them into sub-classes, and compares evaluation approaches.
Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.
citing papers explorer
-
Jailbreak Attacks and Defenses Against Large Language Models: A Survey
A survey that creates taxonomies for jailbreak attacks and defenses on LLMs, subdivides them into sub-classes, and compares evaluation approaches.