AIBearisharXiv – CS AI · 4h ago7/10
🧠
Erased, but Not Gone: Output Forgetting Is Not True Forgetting
Researchers demonstrate that machine unlearning methods that appear successful at the output layer—the standard evaluation metric—actually retain structured residual information in representation space compared to true retraining. This finding reveals a critical gap between apparent forgetting and genuine forgetting, suggesting current unlearning evaluations systematically overestimate effectiveness.