FedUP achieves fast, reversible one-shot federated unlearning via pluggable centroid-guided filters that reduce non-target knowledge loss.
Exact unlearning of finetuning data via model merging at scale.arXiv preprint arXiv:2504.04626
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
LLM unlearning is reframed as inadvertently installing backdoor triggers on forget-tokens; Random Noise Augmentation is introduced as a defense that improves robustness with theoretical guarantees.
citing papers explorer
-
FedUP: One-Shot Federated Unlearning via Centroid-Guided Plug-in Filters
FedUP achieves fast, reversible one-shot federated unlearning via pluggable centroid-guided filters that reduce non-target knowledge loss.
-
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
-
Improving LLM Unlearning Robustness via Random Perturbations
LLM unlearning is reframed as inadvertently installing backdoor triggers on forget-tokens; Random Noise Augmentation is introduced as a defense that improves robustness with theoretical guarantees.