mlsanityData to 7 Oct 2026

Contextual Memory Virtualisation: DAG-Based State Management and Structurally Lossless Trimming for LLM Agents

arXiv 2602.22402 ↗First posted 25 Feb 2026cs.SE cs.AI cs.HCcited by 0

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

As large language models engage in extended reasoning tasks, they accumulate significant state -- architectural mappings, trade-off decisions, codebase conventions -- within the context window. This understanding is lost when sessions reach context limits and undergo lossy compaction. We propose Contextual Memory Virtualisation (CMV), a system that treats accumulated LLM understanding as version-controlled state. Borrowing from operating system virtual memory, CMV models session history as a Directed Acyclic Graph (DAG) with formally defined snapshot, branch, and trim primitives that enable context reuse across independent parallel sessions. We introduce a three-pass structurally lossless trimming algorithm that preserves every user message and assistant response verbatim while reducing token counts by a mean of 20% and up to 86% for sessions with significant overhead by stripping mechanical bloat such as raw tool outputs, base64 images, and metadata. A single-user case-study evaluation across 76 real-world coding sessions demonstrates that trimming remains economically viable under prompt caching, with the strongest gains in mixed tool-use sessions, which average 39% reduction and reach break-even within 10 turns. A reference implementation is available at https://github.com/CosmoNaught/claude-code-cmv.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 8 of its 10 references that are in the map. First posted on arXiv 25 Feb 2026.

Builds on

8 of 10 references.

References · 8

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,160
  3. 2023MemGPT: Towards LLMs as Operating Systemscited by 1,295
  4. 2023LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compressioncited by 443
  5. 2023Adapting Language Models to Compress Contextscited by 350
  6. 2023RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentationcited by 253
  7. 2023In-context Autoencoder for Context Compression in a Large Language Modelcited by 190
  8. 2024AIOS: LLM Agent Operating Systemcited by 94

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Persistent Context Graphs for Efficient Memory Compaction in LLM Agents
  2. 2026TokenPilot: Cache-Efficient Context Management for LLM Agents
  3. 2026Shared Selective Persistent Memory for Agentic LLM Systems
  4. 2026LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioception
  5. 2026Akashic: A Low-Overhead LLM Inference Service with MemAttention
  6. 2026ContextWeaver: Selective and Dependency-Structured Memory Construction for LLM Agents
  7. 2026Memory as Infrastructure: Reliability Engineering for Persistent Agent Memory in Months-Long LLM-Assisted Development
  8. 2026Multi-Head Recurrent Memory Agents