[KongchangAI]
Benchmark

RWKU

An unlearning evaluation benchmark based on real-world person knowledge, used to test the robustness of large-model unlearning and to better approximate practical deployment scenarios.

Timeline (last 90 days)

Nov 17

LEAF-OPD's forget quality on the RWKU benchmark is 34% higher than the second-place method, while retain quality remains competitive.

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