mlsanityData to 14 Oct 2026

MemTool: Optimizing Short-Term Memory Management for Dynamic Tool Calling in LLM Agent Multi-Turn Conversations

arXiv 2507.21428 ↗First posted 29 Jul 2025cs.CLcited by 21

On the map: LLM agents and tool use → Tool learning and function calling

Abstract (arXiv)

Large Language Model (LLM) agents have shown significant autonomous capabilities in dynamically searching and incorporating relevant tools or Model Context Protocol (MCP) servers for individual queries. However, fixed context windows limit effectiveness in multi-turn interactions requiring repeated, independent tool usage. We introduce MemTool, a short-term memory framework enabling LLM agents to dynamically manage tools or MCP server contexts across multi-turn conversations. MemTool offers three agentic architectures: 1) Autonomous Agent Mode, granting full tool management autonomy, 2) Workflow Mode, providing deterministic control without autonomy, and 3) Hybrid Mode, combining autonomous and deterministic control. Evaluating each MemTool mode across 13+ LLMs on the ScaleMCP benchmark, we conducted experiments over 100 consecutive user interactions, measuring tool removal ratios (short-term memory efficiency) and task completion accuracy. In Autonomous Agent Mode, reasoning LLMs achieve high tool-removal efficiency (90-94% over a 3-window average), while medium-sized models exhibit significantly lower efficiency (0-60%). Workflow and Hybrid modes consistently manage tool removal effectively, whereas Autonomous and Hybrid modes excel at task completion. We present trade-offs and recommendations for each MemTool mode based on task accuracy, agency, and model capabilities.

Key facts. Cited by 21 papers in the map (data through 14 Oct 2026). Placed in Tool learning and function calling, one of 16 directions in LLM agents and tool use. Builds on 25 of its 49 references that are in the map. First posted on arXiv 29 Jul 2025.

Builds on

25 of 49 references.

Closest ancestors · 9

Same direction, newest first

  1. 2025MCP-Zero: Active Tool Discovery for Autonomous LLM Agents1 Jun
  2. 2025ScaleMCP: Dynamic and Auto-Synchronizing Model Context Protocol Tools for LLM Agents9 May
  3. 2025Evaluating Personalized Tool-Augmented LLMs from the Perspectives of Personalization and Proactivity2 Mar
  4. 2024Toolshed: Scale Tool-Equipped Agents with Advanced RAG-Tool Fusion and Tool Knowledge Bases18 Oct
  5. 2024CITI: Enhancing Tool Utilizing Ability in Large Language Models without Sacrificing General Performance20 Sep
  6. 2024Seal-Tools: Self-Instruct Tool Learning Dataset for Agent Tuning and Detailed Benchmark14 May
  7. 2024AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls6 Feb
  8. 2023Gorilla: Large Language Model Connected with Massive APIs24 May
  9. 2023ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings19 May

Other references · 16

  1. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  2. 2023MemGPT: Towards LLMs as Operating Systemscited by 1,389
  3. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,063
  4. 2024Evaluating Very Long-Term Conversational Memory of LLM Agentscited by 830
  5. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 707
  6. 2025LLMs Get Lost In Multi-Turn Conversationcited by 429
  7. 2024Graph Retrieval-Augmented Generation: A Surveycited by 324
  8. 2025Agentic Retrieval-Augmented Generation: A Survey on Agentic RAGcited by 249
  9. 2023CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsetscited by 129
  10. 2025From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMscited by 124
  11. 2025On Memory Construction and Retrieval for Personalized Conversational Agentscited by 105
  12. 2024Re-Invoke: Tool Invocation Rewriting for Zero-Shot Tool Retrievalcited by 52
4 more
  1. 2024ToolRerank: Adaptive and Hierarchy-Aware Reranking for Tool Retrievalcited by 41
  2. 2025LongRoPE2: Near-Lossless LLM Context Window Scalingcited by 24
  3. 2023ProTIP: Progressive Tool Retrieval Improves Planningcited by 12
  4. 2025Hephaestus: Improving Fundamental Agent Capabilities of Large Language Models through Continual Pre-Trainingcited by 6

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 8; 2026: 13.

21 papers.

Sort:

By area:

  1. 2025Memory in the Age of AI Agentscited by 303
  2. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memorycited by 284
  3. 2026A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agentscited by 44
  4. 2025A Survey on Agentic Multimodal Large Language Modelscited by 23
  5. 2026Is Grep All You Need? How Agent Harnesses Reshape Agentic Searchcited by 21
  6. 2026Self-Improvements in Modern Agentic Systems: A Surveycited by 19
  7. 2026Mem-T: Densifying Rewards for Long-Horizon Memory Agentscited by 18
  8. 2025Tool-to-Agent Retrieval: Bridging Tools and Agents for Scalable LLM Multi-Agent Systemscited by 15
  9. 2025AI Meets Brain: Memory Systems from Cognitive Neuroscience to Autonomous Agentscited by 12
  10. 2026Cited but Not Verified: Parsing and Evaluating Source Attribution in LLM Deep Research Agentscited by 12
  11. 2026LMEB: Long-horizon Memory Embedding Benchmarkcited by 7
  12. 2026How Many Tools Should an LLM Agent See? A Chance-Corrected Answercited by 7
9 more
  1. 2025Comparison of Text-Based and Image-Based Retrieval in Multimodal Retrieval Augmented Generation Large Language Model Systemscited by 5
  2. 2026SwiftMem: Fast Agentic Memory via Query-aware Indexingcited by 5
  3. 2025Agent-as-a-Graph: Knowledge Graph-Based Tool and Agent Retrieval for LLM Multi-Agent Systemscited by 4
  4. 2026Beyond Rows to Reasoning: Agentic Retrieval for Multimodal Spreadsheet Understanding and Editingcited by 4
  5. 2026Ask Early, Ask Late, Ask Right: When Does Clarification Timing Matter for Long-Horizon Agents?cited by 4
  6. 2026Memory-Induced Tool-Drift in LLM Agentscited by 4
  7. 2025Rethinking Retrieval: From Traditional Retrieval Augmented Generation to Agentic and Non-Vector Reasoning Systems in the Financial Domain for Large Language Modelscited by 2
  8. 2026Agentic Jackal: Live Execution and Semantic Value Grounding for Text-to-JQLcited by 0
  9. 2026DADL: A Declarative Description Language for Enterprise Tool Libraries in LLM Agent Systemscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026MemGym: a Long-Horizon Memory Environment for LLM Agents
  2. 2026MemToolAgent: Leveraging Memory for Tool Using Agents Based on Environment and User Feedback
  3. 2026Mem2ActBench: A Benchmark for Evaluating Long-Term Memory Utilization in Task-Oriented Autonomous Agents
  4. 2026Stateful Inference for Low-Latency Multi-Agent Tool Calling
  5. 2026Akashic: A Low-Overhead LLM Inference Service with MemAttention
  6. 2026The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration
  7. 2026Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers
  8. 2026UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents