2 citations · 3 across the 6 of their papers we have counts for
Showing cs.LGShow all
3 papers · 1 filter
cs.LG2025
Collapse of Irrelevant Representations (CIR) Ensures Robust and Non-Disruptive LLM Unlearning
Filip Sondej, Yushi Yang
Current unlearning and safety training methods consistently fail to remove dangerous knowledge from language models. We identify the root cause - unlearning targets representations…
cs.LG2025
Robust LLM Unlearning with MUDMAN: Meta-Unlearning with Disruption Masking And Normalization
Filip Sondej, Yushi Yang, Mikołaj Kniejski +1
Language models can retain dangerous knowledge and skills even after extensive safety fine-tuning, posing both misuse and misalignment risks. Recent studies show that even speciali…
cs.LG2024★ 2 cited
How Does DPO Reduce Toxicity? A Mechanistic Neuron-Level Analysis
Yushi Yang, Filip Sondej, Harry Mayne +2
Safety fine-tuning algorithms reduce harmful outputs in language models, yet their mechanisms remain under-explored. Direct Preference Optimization (DPO) is a popular choice of alg…