Mitigating Sequential Reappearance in Diffusion Data-Point Unlearning

2026-09-21

AI Summary

研究解决了扩散数据点遗忘中的顺序重现问题，通过引入目标级评估协议来跟踪每个目标是否被立即遗忘、最终保持遗忘或在后续删除中重现。

Abstract

Diffusion data-point unlearning is typically evaluated immediately after each deletion, even though subsequent requests may repeatedly update the same model. We identify sequential reappearance, a failure mode in which an instance that is initially judged to be forgotten later returns to the memorized regime without reuse of the deleted data or adversarial fine-tuning. To capture this behavior, we introduce a target-level evaluation protocol that tracks whether each target is forgotten immediately, remains forgotten at the end of the sequence, or reappears during subsequent deletions. We further find that targets that later reappear exhibit sharper local denoising-loss geometry after deletion than targets that remain forgotten.

Problem

• sequential reappearance
• diffusion data-point unlearning
• memorized regime

Innovation

• sequential reappearance
• target-level evaluation protocol
• local denoising-loss geometry