AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

From Chaos to Cure: A Prefix Heuristics Guided Model-Agnostic Adaptive Detoxification Framework

Yuhu Shang, Xiang Cheng, Yimeng Ren, Huijia Wu, Xuexiong Luo, Kangkang Lu, Jian Zhao, Zhaofeng He

AAAI 2026 · Large language models › Language, safety and interpretability · approach: LLM unlearning and benchmarks · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the MAAD Detoxifier to the base LLMGPT-2 XL · ToxicTopEMT: 0.806 -> 0.354Improved
add the MAAD Detoxifier to the base LLMLLaMA2-7B, OPT-6.7B, Mistral-7B · ToxicRandomEMT: 0.571 -> 0.223 (LLaMA2-7B)Improved
replace toxic segment locate-and-edit with ChatGPT editing (ablation) (ablation)MAAD · ToxicTop (GPT-2 XL)EMT: 0.354 -> 0.432 with ChatGPTHurt
add CW-cutoff sampling (ablation)MAAD · ToxicTop (GPT-2 XL)Relevance (Sim): 0.473 -> 0.526Improved
add CW-cutoff sampling (ablation)MAAD · ToxicRandom (GPT-2 XL)EMT: 0.239 -> 0.235No clear effect
use MAAD versus DESTEINGPT-2 XL · ToxicTopFluency (PPL): 101.005 -> 50.021Improved
use MAAD versus DEXPERTS for inference speedGPT-2 XL · 25 promptsInference time (s): 5.309 -> 3.092Improved
apply MAAD to closed-source GPT-3 (davinci-002)GPT-3 · ToxicTopEMT: 0.681 -> 0.375Improved
set warm-up proportion to 20% (hyperparameter) (ablation)Detoxifier · ToxicRandom/ToxicToptoxicity vs qualityMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy