abliteration
nounEtymology Blend of ablate + obliteration, see abliterate.
1
Computing, Engineering, Mathematics, Natural sciences, Physical sciences, Sciences Neologism The process of uncensoring a large language model by modifying internal functions to eliminate refusal behaviors while preserving the remaining functions of the model.
Artificial intelligence
- We applied abliteration to Daredevil-8B to uncensor it, which also degraded the model's performance.2024 June 13, Maxime Labonne, “Uncensor any LLM with abliteration”, in Hugging Face Blog, retrieved 29 May 2025:
- Related terms
- abliterate