Cite
Notes
Only stored in your browser.
Attribution
Robust LLM Unlearning with MUDMAN: Meta-Unlearning with Disruption Masking And Normalization
arXiv 2025
Beyond Toxic Neurons: A Mechanistic Analysis of DPO for Toxicity Reduction
arXiv 2024
from 2 papers
Yushi Yang
Adam Mahdi
Harry Mayne
Marcel Windys
Mikołaj Kniejski