New paper rethinks how machine unlearning claims are verified
A new arXiv paper argues that machine unlearning claims need evidence that models have removed the influence of specified training data, addressing the lack of widely deployable post hoc verifiers.
Read the full story at arXiv cs.LG ↗
Timeline · 1 report
- 2026-11-17 05:00 · arXiv cs.LG
Rethinking the Verification of Machine Unlearning