Entity-aligned sampling (MELU) is stabler than 1:1 or cyclic retain-set sampling for LLM unlearning, but the paper's diverse-neighbor claim is contradicted by its own Balanced results.
In: First Conference on Language Modeling (2024), https://openreview.net/forum?id=B41hNBoWLo
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Standard vs. Modular Sampling: Best Practices for Reliable LLM Unlearning
Entity-aligned sampling (MELU) is stabler than 1:1 or cyclic retain-set sampling for LLM unlearning, but the paper's diverse-neighbor claim is contradicted by its own Balanced results.