PrivUn shows privacy unlearning in LLMs produces gradient-driven ripple effects and only shallow forgetting across layers, with new strategies proposed for deeper removal.
The Twelfth International Conference on Learning Representations , year=
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
ETW uses predictive entropy as a proxy for token informativeness to improve selective unlearning in LLMs, achieving better forgetting with less utility loss than prior token-level methods.
citing papers explorer
-
PrivUn: Unveiling Latent Ripple Effects and Shallow Forgetting in Privacy Unlearning
PrivUn shows privacy unlearning in LLMs produces gradient-driven ripple effects and only shallow forgetting across layers, with new strategies proposed for deeper removal.
-
Forget What Matters, Keep the Rest: Selective Unlearning of Informative Tokens
ETW uses predictive entropy as a proxy for token informativeness to improve selective unlearning in LLMs, achieving better forgetting with less utility loss than prior token-level methods.
- Less is More: Geometric Unlearning for LLMs with Minimal Data Disclosure