MIT CSAIL Explains: Recursive Language Models - YouTube
Alex explores Recursive Language Models (RLMs), a framework where language models call one another within a coding environment to tackle complex, long-horizon tasks. This approach shifts from traditional, linear prompting to a recursive structure, enabling systems to handle near-infinite contexts and execute autonomous reasoning without requiring additional model training. Follow along using the transcript. Alex explores Recursive Language Models (RLMs), a framework where language models call one another within a coding environment to tackle complex, long-horizon tasks. This approach shifts from traditional, linear prompting to a recursive structure, enabling systems to handle near-infinite contexts and execute autonomous reasoning without requiring additional model training. Follow along using the transcript.
Alex Zhang, Graduate Student | MIT EECS CSAIL https://alexzhang13.github.io/blog/2025/rlm/ DIrector: Rachel Gordon Preditor: Tim Malieckal Script Supervisor: Alex Shipps Chapters: 00:00 - Introduction 00:27 - LMs are more capable than we’ve realized 01:27 - What is a Recursive Language Model? 02:41 - What does your approach do that people are missing? 04:00 - What does treating a prompt as an environment enable? 05:36 - How does a RLM know what is worth looking at, and what isn’t? 07:30 - What does it take to build a RLM? 08:29 - An example of what a RLM can achieve that was out of reach…
saved by
related reading
- Recursive Language Models | Alex L. Zhangalexzhang13.github.io
- RLMs are Compositional Generalizers with Alex Zhang from MITyoutube.com
- Recursive Language Models — Alex Zhang, MIT PhDyoutube.com
- Recursive Language Models ("RLMs")whynowtech.substack.com
- Lil'Loglilianweng.github.io
- LLM Powered Autonomous Agents | Lil'Loglilianweng.github.io
- Recursive Language Models: the paradigm of 2026primeintellect.ai
- Alex L. Zhangalexzhang13.github.io
- Language Models, World Models, and Human Model-Buildinglingo.csail.mit.edu
- Productizing Large Language Modelsblog.replit.com
- [2511.08579] Training Language Models to Explain Their Own Computationsarxiv.org
- Language Models can Solve Computer Tasksarxiv.org