NUWA 风险治理 · 研究
Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs
NUWA 研究解读 · 独立复核待完成
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联系研究与决策意义证据补充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
预训练数据整理提升了知识代理指标上的抗篡改性，但外部信息可得时仍有重要边界。
研究问题
在预训练中限制领域知识，能否比单独后训练形成更持久防护？
研究设计与假设
作者以匹配架构比较过滤与未过滤语料训练，分别评测通用与代理知识，并测试有界篡改；本次排除危险任务内容。
来源发现
过滤模型降低目标代理知识，同时保留已测无关能力。
论文明确指出上下文信息改变防护边界，没有把任何受测防御称为普遍稳健。
NUWA 解读
NUWA 解读：内部知识代理上的抗性支持一层防御，不证明已防止工具增强的滥用。
局限与相反证据
模型规模、代理构念、训练分布与测试对手资源限定结果，未直接测量真实有害后果。
研究联系
NUWA 关联：开放权重风险议程将数据整理与篡改评测列为不同研究缺口；RAND 观点质疑代理到后果的另一链路。
研究与决策意义
NUWA 建议：防护主张同时披露代理有效性与外部信息假设，扩大部署推断前需独立安全审阅。
精确来源位置
- Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs · 2 Filtering Prevents Target Capabilities: evaluation designSelected original resource https://arxiv.org/html/2508.06601; sha256:e4e1e635d31f067b887ac9777737d130ae3c48052395a265d4fea00bfe4ab8cf; archived source capture; exact capture time not independently established
- Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs · 3 Filtering Achieves State-of-the-Art Tamper-Resistance: aggregate resultsSelected original resource https://arxiv.org/html/2508.06601; sha256:e4e1e635d31f067b887ac9777737d130ae3c48052395a265d4fea00bfe4ab8cf; archived source capture; exact capture time not independently established
- Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs · 4 Defense-in-Depth: reported limitationsSelected original resource https://arxiv.org/html/2508.06601; sha256:e4e1e635d31f067b887ac9777737d130ae3c48052395a265d4fea00bfe4ab8cf; archived source capture; exact capture time not independently established
- Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs · 6.3 LimitationsSelected original resource https://arxiv.org/html/2508.06601; sha256:e4e1e635d31f067b887ac9777737d130ae3c48052395a265d4fea00bfe4ab8cf; archived source capture; exact capture time not independently established
- Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs · 6.4 Impact and Risk StatementSelected original resource https://arxiv.org/html/2508.06601; sha256:e4e1e635d31f067b887ac9777737d130ae3c48052395a265d4fea00bfe4ab8cf; 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 →
证据补充 · 2026-10-28
模型辅助整理 · 独立证据及双语编辑复核待完成。以下数值为来源发现或材料计数，与趋势结果序列分开保存。
防护初始效果、选择性和后续适应下的持久性需要分别核查。蛋白模型过滤与工具重建补充安全论证中的不同前提，不是 Deep Ignorance 的同模型复现；代理分数不等于实际生物危害。
关联阅读
NUWA 阅读关联；不据此声称原文引用、实验复现或成绩等价。
本次解释的来源
- ReproBAIT: Can coding agents build their own biological AI tools? · Abstract; Table2 caption; §3; limitations; text lines 00207Original resource captured 2026-10-28T07:25:52.920583+00:00; sha256:cdc5d63c54b48afc059393a7b154a5670ae73b72361536695ada5890520234f1
- The Effectiveness and Durability of Pretraining Data Filtering in Protein Language Models · Abstract; §4.4 limitations; conclusion; text lines 00271Original resource captured 2026-10-28T07:25:52.921030+00:00; sha256:83a3c96f121b7de5bcb6ba87f388d6ac39e33416fd00b70cbcf68d3235ac7670
机构与角色
作者 (10)
Kyle O'Brien · Stephen Casper · Quentin Anthony · Tomek Korbak · Robert Kirk · Xander Davies · Ishan Mishra · Geoffrey Irving · Yarin Gal · Stella Biderman
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。
原始出版物 · BibTeX
@misc{deep-ignorance,
title = {Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs},
url = {https://arxiv.org/abs/2508.06601},
author = {Kyle O'Brien and Stephen Casper and Quentin Anthony and Tomek Korbak and Robert Kirk and Xander Davies and Ishan Mishra and Geoffrey Irving and Yarin Gal and Stella Biderman},
year = {2025},
date = {2025-08-08},
version = {arXiv:2508.06601v2},
note = {arXiv:2508.06601v2}
}
NUWA 记录 · BibTeX
@misc{deep-ignorance-nuwa,
title = {Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs — NUWA research record},
url = {https://whitzard.tech/nuwa/whitzard-index/work/deep-ignorance/},
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--deep-ignorance.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--deep-ignorance.json.}
}
证据快照 2026-10-29.1 · 索引更新 · RSS
参考文献
- [1] Kyle O’Brien, Stephen Casper 等. Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs. 2025-08-08. v2 [arXiv]; Author affiliations on the original paper title page; 2 Filtering Prevents Target Capabilities: evaluation design; 3 Filtering Achieves State-of-the-Art Tamper-Resistance: aggregate results