Large language models can learn and generalize steganographic chain-of-thought under process supervision

MATS 学员:

Robert McCarthy, Vasil Georgiev

作者:

Joey Skaf, Luis Ibanez-Lissen, Robert McCarthy, Connor Watts, Vasil Georgiv, Hannes Whittingham, Lorena Gonzalez-Manzano, David Lindner, Cameron Tice, Edward James Young, Puria Radmard

引用次数

19 Citations

摘要:

Chain-of-thought (CoT) reasoning not only enhances large language model performance but also provides critical insights into decision-making processes, marking it as a useful tool for monitoring model intent and planning. However, recent works have shown that banning the mention of a specific example of reward hacking causes obfuscation of the undesired reasoning traces but the persistence of the undesired behavior, threatening the reliability of CoT monitoring. We provide an extension to these results with regard to the ability of models to learn a specific type of obfuscated reasoning: steganography. First, we show that penalizing the use of specific strings within load-bearing reasoning traces causes models to substitute alternative strings. Crucially, this does not alter the underlying method by which the model performs the task, demonstrating that the model can learn to steganographically encode its reasoning.We further demonstrate that models can generalize an encoding scheme. When the penalized strings belong to an overarching class, the model learns not only to substitute strings seen in training, but also develops a general encoding scheme for all members of the class which it can apply to held-out testing strings.

近期研究

非大国冲突与 AI 风险

作者:

Kristina Kempkey

日期:

八月 26, 2026

引用次数:

合成角色预训练:从第一个 token 开始对齐

作者:

Julian Minder

日期:

八月 13, 2026

引用次数:

常见问题

什么是 MATS 项目?
项目持续多久?