UCD steers an LLM away from forget-set content at inference time by mixing in the difference between forget-tuned and retain-tuned small models.
Regulation (eu) 2016/679 of the european parliament and of the council of 27 april
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
UCD: Unlearning in LLMs via Contrastive Decoding
UCD steers an LLM away from forget-set content at inference time by mixing in the difference between forget-tuned and retain-tuned small models.