一作ACL 2026Poster· Natural Language Processing
REMIND : Memorization and Unlearning in LLM s Through the Lens of Input Loss Landscapes
Liran Cohen · Yaniv Nemcovsky · Avi Mendelson
摘要
Understanding how large language models (LLMs) store, retain, and remove knowledge is critical for interpretability, reliability, and privacy compliance. We reveal a key phenomenon: machine unlearning imprints distinct geometric signatures in the model’s input loss landscape (ILL), with unlearned...