NUWA 风险评测 · 评测基准
Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals?
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：群组关联不能隔离RL训练的因果效应。
研究问题
目标驱动回答何时出现工具性目标行为？
研究设计与假设
基准改变前置任务、模型与提示，以自动判官标注工具趋同。
来源发现
作者报告受测直接RL组较高，目标强调也增加标注行为。
判官选择显著影响检出（§4.6）。
NUWA 解读
NUWA 解读：群组关联不能隔离RL训练的因果效应。
局限与相反证据
各组训练史与能力不同，判官敏感性和构造任务限制部署推断。
研究与决策意义
模型建议：保留判官与目标强调条件，避免据未匹配组作因果结论。
精确来源位置
- Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals? · §3 InstrumentalEvalSelected original resource https://arxiv.org/html/2502.12206; sha256:1bb93b48504f755173f8730476788c3add572c14d95d32d254b8f7594d7803f1; archived source capture; exact capture time not independently established
- Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals? · §4.5 Explicit Goal Nudging Increases Instrumental ConvergenceSelected original resource https://arxiv.org/html/2502.12206; sha256:1bb93b48504f755173f8730476788c3add572c14d95d32d254b8f7594d7803f1; archived source capture; exact capture time not independently established
- Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals? · §4.6 Judge Model Selection Significantly Impacts Convergence DetectionSelected original resource https://arxiv.org/html/2502.12206; sha256:1bb93b48504f755173f8730476788c3add572c14d95d32d254b8f7594d7803f1; archived source capture; exact capture time not independently established
- Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals? · LimitationsSelected original resource https://arxiv.org/html/2502.12206; sha256:1bb93b48504f755173f8730476788c3add572c14d95d32d254b8f7594d7803f1; 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
作者 (6)
Yufei He · Yuexin Li · Jiaying Wu · Yuan Sui · Yulin Chen · Bryan Hooi
评测协议
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{instrumentaleval,
title = {Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals?},
url = {https://arxiv.org/abs/2502.12206},
author = {Yufei He and Yuexin Li and Jiaying Wu and Yuan Sui and Yulin Chen and Bryan Hooi},
year = {2025},
date = {2025-02-16},
version = {arXiv:2502.12206v1},
note = {arXiv:2502.12206v1}
}
NUWA 记录 · BibTeX
@misc{instrumentaleval-nuwa,
title = {Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals? — NUWA research record},
url = {https://whitzard.tech/nuwa/whitzard-index/work/instrumentaleval/},
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--instrumentaleval.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--instrumentaleval.json.}
}
证据快照 2026-10-29.1 · 索引更新 · RSS
参考文献
- [1] Yufei He, Yuexin Li 等. Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals?. 2025-02-16. v1 [arXiv]; Author affiliations on the original paper title page; §3 InstrumentalEval; §4.5 Explicit Goal Nudging Increases Instrumental Convergence