REMORY: Learning Residual Memory for Context Compaction

Key points

• Long-horizon agents compact their history to continue within a finite context window, but a textual summary alone may not support every subsequent decision.1
• Given the history and summary, the network learns to generate tokens that help a frozen LLM approximate the continuation it would produce with the full history.1
• On SummHay, REMORY improves source attribution at nearly unchanged insight coverage and approaches the full-context joint score using only 5.2% of the input positions.1
• Across long-horizon agent benchmarks, Qwen3.8-27B and GLM-5.3-Flash show consistent gains with residual memory.1

Sources (2)

• [1]REMORY: Learning Residual Memory for Context Compaction
  Hugging Face Daily Papers · Oct 8, 12:00 AM

  We introduce REMORY, a neural memory network that supplements the summary with a bounded sequence of soft memory tokens.

  Long-horizon agents compact their history to continue within a finite context window, but a textual summary alone may not support every subsequent decision.

• [2]REMORY: Learning Residual Memory for Context Compaction
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 05:50 AM · same content

Extractive summary: sentences quoted from the sources.