CLM author responds to RLM debate: context as a variable, and the two can combine into stronger agents
RulinShao · x · 2026-10-01
- Rulin Shao responds to lateinteraction's debate on Context Language Models (CLMs) vs RLMs, noting CLMs were inspired by RLMs: treating context as a variable and interacting with it via a general programming language.
- In the quoted tweet, lateinteraction argues an RLM could express the same behavior (everything is Turing-complete) but carries different inductive biases: the RLM root model operates sequentially and accumulates context, whereas in CLMs the whole context is editable.
- The author sees them as complementary: CLMs control a live, editable context while RLMs handle long inputs, and combining both could yield more powerful agents.
More from coding & agent
- Mathematician details agentic math workflow with Codex CLI and Claude Code, results coming — burny_tech · 2026-10-01
- Fathom: An Individuality System for Agents That Matches Top Memory Systems — allisonmaybe · 2026-10-01
- Airbench Crowdsources a Local LLM Leaderboard via One-Prompt Agent Benchmarks — dh7net · 2026-10-01
- Vercel Ship SF Agenda: Notion's Agentic Platform, Grok Powering 300K Apps, Vercel's Eve Agent Framework — evilrabbit_ · 2026-10-01
- Azure AI kicks off series: why content extraction matters more as GenAI models get stronger — adnan_hashmi · 2026-10-01
- Inside OpenAI DevDay: superhuman computer use, Decisions API, and the AI cloud — Latent Space · 2026-10-01