mlsanityData to 14 Oct 2026

CrystalMem: Elastic Memory for Self-Evolving LLM Agents via Knowledge Crystallization

arXiv 2608.00303 ↗First posted 31 Jul 2026cs.AIcited by 2

On the map: LLM agents and tool use → Agent memory systems

also in 1 more
Abstract (arXiv)

Memory for self-evolving large language model (LLM) agents is often provisioned as if its byte budget only grows. Cloud platforms, however, adjust quotas with load and cost, and we show that capability does not follow the budget back up: after a squeeze-and-recover cycle, the agent settles below its pre-squeeze level, a gap we call memory hysteresis. The cause is structural. Deletion and one-way compression discard the material needed for later rebuilding, and we prove that any policy that only keeps or drops entries carries a residual-deficit floor. We propose CrystalMem (Crystallized Memory), an elastic memory sidecar that demotes entries across four fidelity states under a crystallization-energy schedule, orders demotions by advantage-weighted influence with dependency coupling, and recovers capability through verified recrystallization under explicit compute and byte caps. Across seven environments, seventeen methods, and six backbones, with multi-tenant serving and a physical edge-cloud deployment, CrystalMem achieves the highest restored capability in every setting and closes the loop left open by every baseline. From a 50% byte budget, CrystalMem matches the strongest budgeted baseline at full provision on every environment; at equal budgets, it leads by +4.6 pp on average.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Agent memory systems, one of 16 directions in LLM agents and tool use. Builds on 51 of its 72 references that are in the map. First posted on arXiv 31 Jul 2026.

Builds on

51 of 72 references.

Closest ancestors · 16

Same direction, newest first

  1. 2026What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents9 Jul
  2. 2026Governed Shared Memory for Multi-Agent LLM Systems23 Jun
  3. 2026RecMem: Recurrence-based Memory Consolidation for Efficient and Effective Long-Running LLM Agents15 May
  4. 2026Human-Inspired Memory Architecture for LLM Agents8 May
  5. 2026Experience Compression Spectrum: Unifying Memory, Skills, and Rules in LLM Agents17 Apr
  6. 2026Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers8 Mar
  7. 2026Adaptive Memory Admission Control for LLM Agents4 Mar
  8. 2025Forgetful but Faithful: A Cognitive Memory Architecture and Benchmark for Privacy-Aware Generative Agents14 Dec
  9. 2025Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks14 Oct
  10. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory29 Sep
  11. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactions7 Jul
  12. 2025MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents18 Jun
4 more
  1. 2025Memory OS of AI Agent30 May
  2. 2025Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory28 Apr
  3. 2025A-MEM: Agentic Memory for LLM Agents17 Feb
  4. 2024Evaluating Very Long-Term Conversational Memory of LLM Agents27 Feb

Other references · 35

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  3. 2024Qwen2.5 Technical Reportcited by 4,942
  4. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,191
  5. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 2,006
  6. 2023MemGPT: Towards LLMs as Operating Systemscited by 1,389
  7. 2020ALFWorld: Aligning Text and Embodied Environments for Interactive Learningcited by 1,205
  8. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,073
  9. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 707
  10. 2024Agent Workflow Memorycited by 383
  11. 2018BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learningcited by 244
  12. 2025EvolveR: Self-Evolving LLM Agents through an Experience-Driven Lifecyclecited by 137
23 more
  1. 2026Harness VLA: Steering Frozen VLAs into Reliable Manipulation Primitives via Memory-Guided Agentscited by 47
  2. 2025Sleep-time Compute: Beyond Inference Scaling at Test-timecited by 43
  3. 2025PolySkill: Learning Generalizable Skills Through Polymorphic Abstractioncited by 27
  4. 2024Prioritized Information Bottleneck Theoretic Framework with Distributed Online Learning for Edge Video Analyticscited by 17
  5. 2024R-ACP: Real-Time Adaptive Collaborative Perception Leveraging Robust Task-Oriented Communicationscited by 13
  6. 2026Inference-Time Budget Control for LLM Search Agentscited by 13
  7. 2025R3Mem: Bridging Memory Retention and Retrieval via Reversible Compressioncited by 12
  8. 2025Enhancing Vehicular Platooning with Wireless Federated Learning: A Resource-Aware Control Frameworkcited by 9
  9. 2026Forget to Improve: On-Device LLM-Agent Continual Learning via Budget-Curated Memorycited by 9
  10. 2025"X of Information'' Continuum: A Survey on AI-Driven Multi-dimensional Metrics for Next-Generation Networked Systemscited by 7
  11. 2025Shared Spatial Memory Through Predictive Codingcited by 7
  12. 2026Darwinian Memory: A Training-Free Self-Regulating Memory System for GUI Agent Evolutioncited by 7
  13. 2026RELIEF: Turning Missing Modalities into Training Acceleration for Federated Learning on Heterogeneous IoT Edgecited by 7
  14. 2026Lifecycle-Aware Federated Continual Learning in Mobile Autonomous Systemscited by 7
  15. 2026Remember the Decision, Not the Description: A Rate-Distortion Framework for Agent Memorycited by 7
  16. 2026Reclaim Evaluation: A Lossy Memory Is Worse Than an Empty Onecited by 7
  17. 2026Reinforcement Learning-Based Energy-Aware Coverage Path Planning for Precision Agriculturecited by 6
  18. 2026Application-Aware Twin-in-the-Loop Planning for Federated Split Learning over Wireless Edge Networkscited by 5
  19. 2026EASE: Federated Multimodal Unlearning via Entanglement-Aware Anchor Closurecited by 5
  20. 2026SCALE: Sensitivity-Aware Federated Unlearning with Information Freshness Optimization for Mobile Edge Computingcited by 5
  21. 2026PRISM: Exposing and Resolving Spurious Isolation in Federated Multimodal Continual Learningcited by 3
  22. 2025Temporal-Aware GPU Resource Allocation for Distributed LLM Inference via Reinforcement Learningcited by 2
  23. 2026Neuromem: A Granular Decomposition of the Streaming Lifecycle in External Memory for LLMscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026ERRAND: Budgeted Maintenance of Agent Memorycited by 1
  2. 2026Not All Experience Belongs in the Weights: Component Routing for Self-Improving GUI Agentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026SAGE: A Novelty Gate for Efficient Memory Evolution in Agentic LLMs
  2. 2026Chain-of-Memory: Lightweight Memory Construction with Dynamic Evolution for LLM Agents
  3. 2026PLACEMEM: Toward a Compute-Aware Memory Plane for Lifelong Agents
  4. 2026Compact-Memory LLM Agents via Online Max-Member Clustering and Atom-Aware Packing
  5. 2026Pancake: Hierarchical Memory System for Multi-Agent LLM Serving
  6. 2026Towards Autonomous Memory Agents
  7. 2026Akashic: A Low-Overhead LLM Inference Service with MemAttention
  8. 2026MemMA: Coordinating the Memory Cycle through Multi-Agent Reasoning and In-Situ Self-Evolution