[2510.27338] Reasoning Models Sometimes Output Illegible Chains of Thought

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status

[2510.27338] Reasoning Models Sometimes Output Illegible Chains of Thought Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Machine Learning arXiv:2510.27338 (cs) [Submitted on 31 Oct 2025] Title: Reasoning Models Sometimes Output Illegible Chains of Thought Authors: Arun Jose View a PDF of the paper titled Reasoning Models Sometimes Output Illegible Chains of Thought, by Arun Jose View PDF HTML (experimental) Abstract: Language models trained via outcome-based reinforcement learnin

related reading

• Tracing the thoughts of a large language model \ Anthropicanthropic.com
• [2505.05410] Reasoning Models Don't Always Say What They Thinkarxiv.org
• Reasoning models don't always say what they think \ Anthropicanthropic.com
• Can Reasoning Models Obfuscate Reasoning? Stress-Testing Chain-of-Thought Monitorabilityarxiv.org
• When Chain of Thought is Necessary, Language Models Struggle to Evade Monitorsarxiv.org
• Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscationarxiv.org
• DeepSeek-R1arxiv.org
• the-illusion-of-thinking.pdfml-site.cdn-apple.com
• How AI Is Learning to Think in Secret — LessWronglesswrong.com
• [2603.05706] Reasoning Models Struggle to Control their Chains of Thoughtarxiv.org
• Vestigial reasoning in RL — LessWronglesswrong.com
• Announcing ReasoningLens — Visualizing and Diagnosing LLM Reasoning at a Glancehuggingface.co