Recursive Language Models
❗please read❗
Participants are to read the material in advance to engage more fully with the technical and methodological details during the session.
If you have your own research, experience or saw something on the internet (blogpost, article etc.) that would add to the discussion, do contribute!
Suggested Pre-Reading:
Main paper: Recursive Language Models
Optional papers: Tree of Thoughts, MemPalace, A-MEM
Language models are increasingly expected to handle inputs that exceed their fixed context windows, whether through longer prompts, larger documents, or more complex multi-step tasks.
Let's examine "Recursive Language Models" one possible approach to that problem. Instead of fitting everything into a single forward pass, RLMs treat the prompt as an external environment and recursively inspect relevant snippets as needed during inference.
The discussion situates RLMs within a broader lineage of approaches for handling long or external information.
RAG for retrieving relevant context on demand,
ReAct for iterative interaction with external information,
MemGPT for explicit OS-style memory management, and
UltraLong for pushing the context window itself to extreme scales through instruction tuning.
Together, these papers reflect different strategies for retrieval, memory management, iterative interaction, and long-context scaling, while RLMs introduce a different direction: recursive inference-time navigation over context.
We'll then also compare RLMs against frontier long-context models and examine the tradeoffs between context length, reasoning ability, compute efficiency, and controllability.
More About the Host
Keshav Nath completed his Master of Computing from the National University of Singapore in 2026, after earning a Bachelor’s degree in Mechanical Engineering from Delhi Technological University. He has experience across both applied AI and research. He was a student researcher at the UCLA Smart Energy Lab and completed the fully funded MITACS Globalink Research Internship onsite in Toronto, resulting in three publications. In Singapore, he worked as a Junior AI Engineer at Staple AI, building production-grade agentic systems, and as a Graduate Teaching Assistant at NUS for Advanced Analytics with Big Data.
More About the Series
Paper Club is Lorong AI’s community-driven initiative where members gather to discuss and analyze academic papers, research articles, or key developments in artificial intelligence.
Get involved: Learn more about Lorong AI | Speaker Sign-up | WhatsApp Community | LinkedIn | X