o1 and Reasoning | AndoLogs

Organizing research presumably related to OpenAI's o1 reasoning model.

Curious about o1 and “reasoning” in LLMs? Then this post is for you! Here, I attempt to aggregate and organize research related to OpenAI’s o1 . This post is heavily inspired by this awesome-o1 repo/talk [ 1 ] and O1-Journey [ 2 ] for sources and insights grounded in the community’s current consensus, but I’ve also added my own thoughts and interpretations. Disclaimer: All of the thoughts presented here are speculative based on only the publicly available information about o1. I will continue to update this post as more information becomes available, but none of it will come from official or c

saved by

• Clementine Yang

related reading

• VRPRM: Process Reward Modeling via Visual Reasoningarxiv.org
• o1: A Technical Primer — LessWronglesswrong.com
• DeepSeek-R1arxiv.org
• Learning to reason with LLMs | OpenAIopenai.com
• the-illusion-of-thinking.pdfml-site.cdn-apple.com
• DeepSeek R1's recipe to replicate o1 and the future of reasoning LMsinterconnects.ai
• The State of Reinforcement Learning for LLM Reasoningmagazine.sebastianraschka.com
• Explore | alphaXivalphaxiv.org
• State of RL for reasoning LLMs | A. Weersaweers.de
• Why reasoning models will generalize - by Nathan Lambertinterconnects.ai
• [2201.11903] Chain of Thought Prompting Elicits Reasoning in Large Language Modelsarxiv.org
• Can activation verbalizers surface an internal chain of thought? — LessWronglesswrong.com