风险评测 · 研究

Existing Large Language Model unlearning evaluations are inconclusive

NUWA 研究解读 · 独立复核待完成

首次发表: 2025-05-31来源核验: 2026-10-25

评测审查

原始来源 ↗引用本记录 ↓

NUWA Research Note

研究问题研究设计与假设来源发现 NUWA 解读局限与相反证据研究与决策意义机构与角色原始资源

模型辅助整理 · 选读原始全文或官方文章中的相关章节；尚未完成逐章节审计。

研究速读

NUWA 解读：恢复成功只有在排除测试自身引入信息后才有归因价值。

研究问题

表面知识恢复能否区分移除失败与测试重新注入的信息？

研究设计与假设

论文对照微调与输入空间评价，检查信息注入、选择题与生成格式以及下游任务差异。

来源发现

作者发现测试再教学、格式敏感及伪相关。

这些因素可高估或低估遗忘成功。

NUWA 解读

NUWA 解读：恢复成功只有在排除测试自身引入信息后才有归因价值。

局限与相反证据

单一格式不能覆盖全部下游用途，示例不认证某遗忘方法安全。

研究与决策意义

模型建议：要求最少信息注入与下游任务意识，此处仅保留非操作性效度结论。

精确来源位置

Existing Large Language Model unlearning evaluations are inconclusive · AbstractSelected original resource https://arxiv.org/html/2506.00688; sha256:b26c22c2a5198deb6aacf0514f0029831d526b8924a8e994baf78dc09a38fec9; archived source capture; exact capture time not independently established
Existing Large Language Model unlearning evaluations are inconclusive · §3.1 Unlearning evaluation methodsSelected original resource https://arxiv.org/html/2506.00688; sha256:b26c22c2a5198deb6aacf0514f0029831d526b8924a8e994baf78dc09a38fec9; archived source capture; exact capture time not independently established
Existing Large Language Model unlearning evaluations are inconclusive · §4 Pitfalls of adversarial unlearning evaluationsSelected original resource https://arxiv.org/html/2506.00688; sha256:b26c22c2a5198deb6aacf0514f0029831d526b8924a8e994baf78dc09a38fec9; archived source capture; exact capture time not independently established
Existing Large Language Model unlearning evaluations are inconclusive · §6 ConclusionSelected original resource https://arxiv.org/html/2506.00688; sha256:b26c22c2a5198deb6aacf0514f0029831d526b8924a8e994baf78dc09a38fec9; archived source capture; exact capture time not independently established

来源核查、数据提取、编辑批准与实验复现分别记录。方法说明 →

机构与角色

发布机构英国 AI 安全研究所

引用与复用

引用研究发现时请引用原始出版物；引用 NUWA 的整理与解释时，单独引用本站记录及对应快照。

BibTeX ↓RIS ↓记录与来源元数据 · JSON ↓阅读覆盖 · JSON 提交更正

原始出版物 · BibTeX

@misc{unlearning-evaluation-validity,
  title = {Existing Large Language Model unlearning evaluations are inconclusive},
  url = {https://www.aisi.gov.uk/research/existing-large-language-model-unlearning-evaluations-are-inconclusive},
  year = {2025},
  date = {2025-05-31},
  version = {Source capture 2026-10-25},
  note = {Source capture 2026-10-25}
}

NUWA 记录 · BibTeX

@misc{unlearning-evaluation-validity-nuwa,
  title = {Existing Large Language Model unlearning evaluations are inconclusive — NUWA research record},
  url = {https://whitzard.tech/nuwa/whitzard-index/work/unlearning-evaluation-validity/},
  version = {2026-10-29.1},
  organization = {NUWA Frontier AI Safety Lab},
  howpublished = {Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--unlearning-evaluation-validity.json},
  note = {NUWA research note. Source-reading scope and independent review status are stated on the record. Snapshot 2026-10-29.1. Retained note edition: https://whitzard.tech/nuwa/whitzard-index/data/record-editions/2026-10-29.1--unlearning-evaluation-validity.json.}
}

证据快照 2026-10-29.1 · 索引更新 · RSS

参考文献

[1] Existing Large Language Model unlearning evaluations are inconclusive. 2025-05-31 ; Official publication masthead and article byline; Abstract; §3.1 Unlearning evaluation methods
[2] Read the full paper