RSM-full paper hits 83% of full-context quality at 32% token cost for memory-limited LLM agents
rohanpaul_ai · x · 2026-09-11
An arXiv paper (2609.04915) introduces RSM-full, an online clustered-memory pipeline for long-horizon LLM agents facing tight prompt budgets. Key idea: organize the past before optimizing retrieval — group related memories and keep them together rather than retrieving isolated chunks.
- Two designs: a cosine-gated max-member merge write rule plus an atom-aware grouped context packer.
- Results: on AMA-Bench it reaches 83% of full-context quality at 32% of token cost under a 4k budget, beating Online K-Means by +3.5–6.0pp (p<.001) across the 2.6k–5k regime. Ablations attribute most gains to the merge rule (+5.7pp) and grouped packer (+5.0pp).
- Replication: improvements reproduce on the independent RealMem benchmark, significantly beating Streaming-Proto (+2.97pp) and Budget-RAG (+0.69pp), roughly on par with BM25-RAG.
Authors: Jiahe Geng, Jinpeng Wang, Kun Yuan.
Related event: RSM Achieves 83% Memory Quality with 32% of Tokens(3 posts)→
More from coding & agent
- Magic Patterns' first engineer: Cursor in Slack, a Patterson review agent, repo clones over worktrees — Y Combinator Blog · 2026-10-28
- Dev builds AI job-search app for his wife, she lands a job, and he open-sources it — CurieuxExplorer · 2026-10-28
- Meshy partners with MCP for Blender: agents drop textured 3D models into your scene — sidahuj · 2026-10-28
- Supercharging LLM classification with logprobs: soft classifiers you can threshold-tune — JnBrymn · 2026-10-28
- Codex daily build series day 23: skipping iCloud plugins for a reset — dotey · 2026-10-28
- Cloudflare Access adds fully customizable pages with HTML + Liquid templates — dinasaur_404 · 2026-10-28