[KongchangAI]
Benchmark

MUSE-News

An unlearning evaluation benchmark focused on the news domain, used to test the unlearning effectiveness of models in news-knowledge scenarios.

Timeline (last 90 days)

Nov 17

LEAF-OPD was evaluated on the MUSE-News benchmark, but the paper abstract does not give specific numerical values.

Unverified50%
Nov 17

TOFU, RWKU and MUSE-News evaluate unlearning methods from three dimensions respectively—synthetic data, real people, and real corpora—and combining the three allows cross-validation of method effectiveness.

Unverified50%

All Facts (2)

Source Articles