Awesome Efficient Agents: A Survey of Memory, Tool Use, and Planning

🤝 **Contributions welcome!** This repository is actively maintained and continuously updated. Feel free to open an issue or submit a pull request to add papers, fix links, or improve categorization.

⚡Introduction

Recent years have seen growing interest in extending large language models into agentic systems. While agent capabilities have advanced rapidly, efficiency has received comparatively less attention despite being crucial for real-world deployment. This repository studies efficiency-guided agent design from three core components: memory, tool use, and planning.

We provide a curated paper list to help readers quickly locate representative work, along with lightweight notes on how each topic connects to efficiency.

• Efficient Memory. We organize memory-related papers into three processes: construction, management, and access.
• Efficient Tool Use. We group papers into tool selection, tool calling, and tool-integrated reasoning.
• Efficient Planning. We collect work on planning that improves overall agent efficiency by reducing unnecessary actions and shortening trajectories.

🧾Paper List

📂 Table of Contents(click to expand/collapse)

• 🧠Memory
  • Latent and Parametric Memory
    • Latent Memory
    • Parametric Memory
  • Textual Memory
    • Prompt-resident Memory
    • Item-based Memory
    • Graph-based Memory
    • Hierarchical Memory
  • Procedural Reuse via Skills
  • Multi-Agent Memory
    • Shared Memory
    • Local Memory
    • Mixed Memory
• 🛠️Tool Use
  • Tool Selection
    • External Retriever
    • Multi-Label Classification (MLC)
    • Vocabulary-based Retrieval
  • Tool Calling
    • In-Place Parameter Filling
    • Parallel Tool Calling
    • Cost-Aware Tool Calling
    • Efficient Test-Time Scaling
    • Efficient Tool Calling with Post-training
  • Tool-Integrated Reasoning (TIR)
    • Selective Invocation
    • Cost-Aware Policy Optimization
• 🧩Planning
  • Single-Agent Planning Efficiency
    • Adaptive Budgeting and Control
    • Structured Search
    • Task Decomposition
    • Policy Optimization
    • Memory and Skill Acquisition
  • Multi-Agent Collaborative Efficiency
    • Topological Efficiency and Sparsification
    • Protocol and Context Optimization
    • Distilling Coordination into Planning
• 📑Related Surveys
  • Memory Survey
  • Tool Use Survey
  • Planning and Reasoning Survey

🧠Memory

In the paper, we organize memory into construction, management, and access. Since many papers overlap across these stages, this README is primarily organized around memory construction to avoid redundancy.

Latent and Parametric Memory

Latent Memory

• (2026-01) FlashMem: Distilling Intrinsic Latent Memory via Computation Reuse
• (2025-09) MemGen: Weaving Generative Latent Memory for Self-Evolving Agents
• (2025-02) M+: Extending MemoryLLM with Scalable Long-Term Memory
• (2025-01) Titans: Learning to Memorize at Test Time
• (2024-09) MemoRAG: Boosting Long Context Processing with Global Memory-Enhanced Retrieval Augmentation
• (2024-07) $\text{Memory}^3$: Language Modeling with Explicit Memory
• (2024-02) MEMORYLLM: Towards Self-Updatable Large Language Models
• (2024-01) Long Context Compression with Activation Beacon

Textual Memory

Prompt-resident Memory

• (2026-07) ACM: Agentic Context Management for Long Horizon Tasks

• (2026-06) Self-Compacting Language Model Agents

• (2026-06) TokenPilot: Cache-Efficient Context Management for LLM Agents

• (2025-10) AgentFold: Long-Horizon Web Agents with Proactive Context Management

• (2025-07) MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent

• (2025-06) MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents

• (2025-04) Dynamic Cheatsheet: Test-Time Learning with Adaptive Memory

• (2024-02) Compress to Impress: Unleashing the Potential of Compressive Memory in Real-World Long-Term Conversations

Item-based Memory

• (2026-08) LeanMem: Simple and Efficient Long-Term Memory for LLM Agents

• (2026-06) MemRefine: LLM-Guided Compression for Long-Term Agent Memory

• (2026-03) Evoking User Memory: Personalizing LLM via Recollection-Familiarity Adaptive Retrieval

• (2026-01) SimpleMem: Efficient Lifelong Memory for LLM Agents

• (2026-01) MemRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic Memory

• (2025-10) Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models

• (2025-09) ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory

• (2025-08) Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning

• (2025-08) Memento: Fine-tuning LLM Agents without Fine-tuning LLMs

• (2025-07) Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving

• (2025-06) Cost-Efficient Serving of LLM Agents via Test-Time Plan Caching(Agentic Plan Caching: Test-Time Memory for Fast and Cost-Efficient LLM Agents)

• (2025-05) From Single to Multi-Granularity: Toward Long-Term Memory Association and Selection of Conversational Agents

• (2025-04) Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory

• (2025-03) MemInsight: Autonomous Memory Augmentation for LLM Agents

• (2025-03) In Prospect and Retrospect: Reflective Memory Management for Long-term Personalized Dialogue Agents

• (2025-02) A-MEM: Agentic Memory for LLM Agents

• (2025-02) On Memory Construction and Retrieval for Personalized Conversational Agents

• (2024-06) Hello Again! LLM-powered Personalized Agent for Long-term Dialogue

• (2024-04) "My agent understands me better": Integrating Dynamic Human-like Memory Recall and Consolidation in LLM-Based Agents

• (2023-10) RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation

• (2023-08) ExpeL: LLM Agents Are Experiential Learners

• (2023-08) MemoChat: Tuning LLMs to Use Memos for Consistent Long-Range Open-Domain Conversation

• (2023-05) MemoryBank: Enhancing Large Language Models with Long-Term Memory

Graph-based Memory

• (2026-07) Zero-Mem: Zero-Token Memory Operations for LLM Agents

• (2026-01) MAGMA: A Multi-Graph based Agentic Memory Architecture for AI Agents

• (2025-10) D-SMART: Enhancing LLM Dialogue Consistency via Dynamic Structured Memory And Reasoning Tree

• (2025-04) Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory

• (2025-01) Zep: A Temporal Knowledge Graph Architecture for Agent Memory

• (2024-07) AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents

• (2024-06) GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models

• (2024-02) KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph

Hierarchical Memory

• (2026-04) StructMem: Structured Memory for Long-Horizon Behavior in LLMs
• (2026-02) Beyond RAG for Agent Memory: Retrieval by Decoupling and Aggregation
• (2026-02) HyMem: Hybrid Memory Architecture with Dynamic Retrieval Scheduling
• (2025-10) Beyond a Million Tokens: Benchmarking and Enhancing Long-Term Memory in LLMs
• (2025-10) LightMem: Lightweight and Efficient Memory-Augmented Generation
• (2025-07) Hierarchical Memory for High-Efficiency Long-Term Reasoning in LLM Agents
• (2025-07) MemOS: A Memory OS for AI System
• (2025-06) Memory OS of AI Agent
• (2024-08) HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model
• (2024-02) A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts
• (2023-10) MemGPT: Towards LLMs as Operating Systems

Procedural Reuse via Skills

• (2026-08) KV-Skill: Forging Expertise in the Model's Native Language

• (2026-05) SkillLens: Adaptive Multi-Granularity Skill Reuse for Cost-Efficient LLM Agents

• (2026-05) SkillOS: Learning Skill Curation for Self-Evolving Agents

• (2026-04) Graph-of-Skills: Dependency-Aware Structural Retrieval for Massive Agent Skills

• (2026-04) SkillClaw: Let Skills Evolve Collectively with Agentic Evolver

• (2026-04) CoEvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification

• (2026-04) SKILLFOUNDRY: Building Self-Evolving Agent Skill Libraries from Heterogeneous Scientific Resources

• (2026-03) SkillReducer: Optimizing LLM Agent Skills for Token Efficiency

• (2026-03) Trace2Skill: Distill Trajectory-Local Lessons into Transferable Agent Skills

• (2026-03) AutoSkill: Experience-Driven Lifelong Learning via Skill Self-Evolution

• (2026-02) MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

• (2026-02) Skill-Pro: Learning Reusable Skills from Experience via Non-Parametric PPO for LLM Agents

• (2026-02) SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning

Multi-Agent Memory

Shared Memory

• (2026-02) LatentMem: Customizing Latent Memory for Multi-Agent Systems
• (2025-11) Latent Collaboration in Multi-Agent Systems
• (2025-11) MemIndex: Agentic Event-based Distributed Memory Management for Multi-agent Systems
• (2025-10) Cache-to-Cache: Direct Semantic Communication Between Large Language Models
• (2025-10) KVCOMM: Online Cross-context KV-cache Communication for Efficient LLM-based Multi-agent Systems
• (2025-08) RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory
• (2025-07) MIRIX: Multi-Agent Memory System for LLM-Based Agents
• (2025-06) G-Memory: Tracing Hierarchical Memory for Multi-Agent Systems
• (2024-04) Memory Sharing for Large Language Model based Agents

Local Memory

• (2025-08) Intrinsic Memory Agents: Heterogeneous Multi-Agent LLM Systems through Structured Contextual Memory
• (2025-04) AgentNet: Decentralized Evolutionary Coordination for LLM-based Multi-Agent Systems
• (2025-02) LLM-Powered Decentralized Generative Agents with Adaptive Hierarchical Knowledge Graph for Cooperative Planning

Mixed Memory

• (2025-10) LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automation
• (2025-05) Collaborative Memory: Multi-User Memory Sharing in LLM Agents with Dynamic Access Control
• (2025-01) SRMT: Shared Memory for Multi-agent Lifelong Pathfinding

🛠️Tool Use

Tool Selection

External Retriever

• (2026-07) Scalable LLM Agent Tool Access in the Cloud

• (2025-10) ToolScope: Enhancing LLM Agent Tool Use through Tool Merging and Context-Aware Filtering

• (2024-10) Toolshed: Scale Tool-Equipped Agents with Advanced RAG-Tool Fusion and Tool Knowledge Bases

• (2024-10) From Exploration to Mastery: Enabling LLMs to Master Tools via Self-Driven Interactions

• (2024-02) AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls

• (2023-12) ProTIP: Progressive Tool Retrieval Improves Planning

Multi-Label Classification (MLC)

• (2024-09) Efficient and Scalable Estimation of Tool Representations in Vector Space
• (2024-09) TinyAgent: Function Calling at the Edge

Vocabulary-based Retrieval

• (2025-03) Chain-of-Tools: Utilizing Massive Unseen Tools in the CoT Reasoning of Frozen Language Models
• (2024-10) Toolken+: Improving LLM Tool Usage with Reranking and a Reject Option
• (2024-10) ToolGen: Unified Tool Retrieval and Calling via Generation
• (2024-07) Concise and Precise Context Compression for Tool-Using Language Models
• (2023-05) ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings

Tool Calling

In-Place Parameter Filling

• (2024-01) Efficient Tool Use with Chain-of-Abstraction Reasoning
• (2023-02) Toolformer: Language Models Can Teach Themselves to Use Tools

Parallel Tool Calling

• (2026-02) W&D:Scaling Parallel Tool Calling for Efficient Deep Research Agents
• (2024-11) CATP-LLM: Empowering Large Language Models for Cost-Aware Tool Planning
• (2024-05) An LLM-Tool Compiler for Fused Parallel Function Calling
• (2023-12) An LLM Compiler for Parallel Function Calling

Cost-Aware Tool Calling

• (2025-07) A Joint Optimization Framework for Enhancing Efficiency of Tool Utilization in LLM Agents
• (2025-05) Distilling LLM Agent into Small Models with Retrieval and Code Tools
• (2025-03) Alignment for Efficient Tool Calling of Large Language Models
• (2025-02) ToolCoder: A Systematic Code-Empowered Tool Learning Framework for Large Language Models
• (2024-02) Budget-Constrained Tool Learning with Planning
• (2024-01) TroVE: Inducing Verifiable and Efficient Toolboxes for Solving Programmatic Tasks

Efficient Test-Time Scaling

• (2026-03) ToolTree: Efficient LLM Agent Tool Planning via Dual-Feedback Monte Carlo Tree Search and Bidirectional Pruning
• (2026-03) Utility-Guided Agent Orchestration for Efficient LLM Tool Use
• (2024-09) ToolPlanner: A Tool Augmented LLM for Multi Granularity Instructions with Path Planning and Feedback
• (2023-10) ToolChain*: Efficient Action Space Navigation in Large Language Models with A* Search

Efficient Tool Calling with Post-training

• (2026-05) On Effectiveness and Efficiency of Agentic Tool-calling and RL Training
• (2025-11) ToolOrchestra: Elevating Intelligence via Efficient Model and Tool Orchestration
• (2025-09) ToolRM: Outcome Reward Models for Tool-Calling Large Language Models
• (2025-04) Acting Less is Reasoning More! Teaching Model to Act Efficiently

Tool-Integrated Reasoning (TIR)

Selective Invocation

• (2025-09) TableMind: An Autonomous Programmatic Agent for Tool-Augmented Table Reasoning
• (2025-05) Tool-Star: Empowering LLM-Brained Multi-Tool Reasoner via Reinforcement Learning
• (2025-02) SMART: Self-Aware Agent for Tool Overuse Mitigation
• (2024-03) Agent-FLAN: Designing Data and Methods of Effective Agent Tuning for Large Language Models

Cost-Aware Policy Optimization

• (2026-03) EvoTool: Self-Evolving Tool-Use Policy Optimization in LLM Agents via Blame-Aware Mutation and Diversity-Aware Selection
• (2026-02) Learning from the Irrecoverable: Error-Localized Policy Optimization for Tool-Integrated LLM Reasoning
• (2026-01) ET-Agent: Incentivizing Effective Tool-Integrated Reasoning Agent via Behavior Calibration
• (2025-10) PORTool: Tool-Use LLM Training with Rewarded Tree
• (2025-10) A$^2$FM: An Adaptive Agent Foundation Model for Tool-Aware Hybrid Reasoning
• (2025-09) Toward Effective Tool-Integrated Reasoning via Self-Evolved Preference Learning
• (2025-09) TableMind: An Autonomous Programmatic Agent for Tool-Augmented Table Reasoning
• (2025-07) Agentic Reinforced Policy Optimization
• (2025-07) AutoTIR: Autonomous Tools Integrated Reasoning via Reinforcement Learning
• (2025-05) Reinforced Internal-External Knowledge Synergistic Reasoning for Efficient Adaptive Search Agent
• (2025-05) Agentic Reasoning and Tool Integration for LLMs via Reinforcement Learning
• (2025-04) ToolRL: Reward is All Tool Learning Needs
• (2025-04) ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
• (2025-04) Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use
• (2025-04) Acting Less is Reasoning More! Teaching Model to Act Efficiently

🧩Planning

Single-Agent Planning Efficiency

Adaptive Budgeting and Control

• (2026-03) Ares: Adaptive Reasoning Effort Selection for Efficient LLM Agents
• (2026-03) SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning
• (2026-02) Think Fast and Slow: Step-Level Cognitive Depth Adaptation for LLM Agents
• (2025-11) Budget-Aware Tool-Use Enables Effective Agent Scaling
• (2025-09) Learning When to Plan: Efficiently Allocating Test-Time Compute for LLM Agents
• (2025-06) Query-Level Uncertainty in Large Language Models
• (2023-12) ReST meets ReAct: Self-Improvement for Multi-Step Reasoning LLM Agent
• (2023-05) SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks
• (2023-03) Reflexion: Language Agents with Verbal Reinforcement Learning

Structured Search

• (2025-05) Cost-Augmented Monte Carlo Tree Search for LLM-Assisted Planning
• (2023-12) ProTIP: Progressive Tool Retrieval Improves Planning
• (2023-10) ToolChain*: Efficient Action Space Navigation in Large Language Models with A* Search
• (2023-10) Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models

Task Decomposition

• (2026-01) Beyond Entangled Planning: Task-Decoupled Planning for Long-Horizon Agents
• (2025-12) Video-Browser: Towards Agentic Open-web Video Browsing
• (2025-05) Alita: Generalist Agent Enabling Scalable Agentic Reasoning with Minimal Predefinition and Maximal Self-Evolution
• (2025-03) ReSo: A Reward-driven Self-organizing LLM-based Multi-Agent System for Reasoning Tasks
• (2024-11) BudgetMLAgent: A Cost-Effective LLM Multi-Agent system for Automating Machine Learning Tasks
• (2024-02) AutoGPT+P: Affordance-based Task Planning with Large Language Models
• (2023-05) ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models
• (2023-03) HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face

Policy Optimization

• (2026-01) WebAnchor: Anchoring Agent Planning to Stabilize Long-Horizon Web Reasoning
• (2025-09) Planner-R1: Reward Shaping Enables Efficient Agentic RL with Smaller LLMs
• (2025-08) Encouraging Good Processes Without the Need for Good Answers: Reinforcement Learning for LLM Agent Planning
• (2025-05) Planning without Search: Refining Frontier LLMs with Offline Goal-Conditioned RL
• (2025-02) QLASS: Boosting Language Agent Inference via Q-Guided Stepwise Search
• (2024-03) Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents

Memory and Skill Acquisition

• (2025-10) GAP: Graph-Based Agent Planning with Parallel Tool Use and Reinforcement Learning
• (2024-07) Sibyl: Simple yet Effective Agent Framework for Complex Real-world Reasoning
• (2024-06) GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models
• (2024-02) Graph-enhanced Large Language Models in Asynchronous Plan Reasoning
• (2023-05) Voyager: An Open-Ended Embodied Agent with Large Language Models

Multi-Agent Collaborative Efficiency

Topological Efficiency and Sparsification

• (2026-04) InfoSeeker: A Scalable Hierarchical Parallel Agent Framework for Web Information Seeking
• (2025-09) MARS: toward more efficient multi-agent collaboration for LLM reasoning
• (2025-08) SafeSieve: From Heuristics to Experience in Progressive Pruning for LLM-based Multi-Agent Communication
• (2025-03) AgentDropout: Dynamic Agent Elimination for Token-Efficient and High-Performance LLM-Based Multi-Agent Collaboration
• (2025-02) S$^2$-MAD: Breaking the Token Barrier to Enhance Multi-Agent Debate Efficiency
• (2024-10) Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems
• (2024-09) GroupDebate: Enhancing the Efficiency of Multi-Agent Debate Using Group Discussion
• (2024-06) Scaling Large Language Model-based Multi-Agent Collaboration
• (2024-06) Chain of Agents: Large Language Models Collaborating on Long-Context Tasks

Protocol and Context Optimization

• • (2026-08) SyncPlan: Long-Horizon LLM Coordination with Explicit Synchronization and Adaptive Correction
• (2026-08) BANDMAS: Causality-Inspired Semantic Packet Scheduling for Bandwidth-Efficient Multi-Agent Collaboration
• (2026-06) Agon: An Autonomous Large-Scale Omnidisciplinary Research System Built on Prompt Economy
• (2025-10) Stop Wasting Your Tokens: Towards Efficient Runtime Multi-Agent Systems
• (2025-09) Free-MAD: Consensus-Free Multi-Agent Debate
• (2025-07) CONSENSAGENT: Towards Efficient and Effective Consensus in Multi-Agent LLM Interactions Through Sycophancy Mitigation
• (2025-07) CodeAgents: A Token-Efficient Framework for Codified Multi-Agent Reasoning in LLMs
• (2024-05) Smurfs: Multi-Agent System using Context-Efficient DFSDT for Tool Planning

Distilling Coordination into Planning

• (2025-11) SMAGDi: Socratic Multi Agent Interaction Graph Distillation for Efficient High Accuracy Reasoning
• (2025-06) Debate, Reflect, and Distill: Multi-Agent Feedback with Tree-Structured Preference Optimization for Efficient Language Model Enhancement
• (2024-02) MAGDi: Structured Distillation of Multi-Agent Interaction Graphs Improves Reasoning in Smaller Language Models

📑Related Surveys

Given that our work mainly focuses on efficiency, which is rooted in effectiveness, we’ve gathered a list of related survey papers to offer a complementary perspective. We hope this will help bring visibility to some valuable surveys that deserve more attention.💡

Memory Survey

• (2026-01) Survey on AI Memory: Theories, Taxonomies, Evaluations, and Emerging Trends
• (2025-12) Memory in the Age of AI Agents
• (2025-05) Rethinking Memory in AI: Taxonomy, Operations, Topics, and Future Directions
• (2025-04) From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs
• (2024-04) A Survey on the Memory Mechanism of Large Language Model based Agents

Tool Use Survey

• (2025-06) LLM-Based Agents for Tool Learning: A Survey
• (2024-05) Tool Learning with Large Language Models: A Survey

Planning and Reasoning Survey

• (2025-08) LLM-based Agentic Reasoning Frameworks: A Survey from Methods to Scenarios
• (2025-02) PlanGenLLMs: A Modern Survey of LLM Planning Capabilities
• (2024-02) Understanding the planning of LLM agents: A survey

📌Citation

If you find this survey useful, please cite:

@misc{yang2026efficientagentsmemorytool,
      title={Toward Efficient Agents: Memory, Tool learning, and Planning}, 
      author={Xiaofang Yang and Lijun Li and Heng Zhou and Tong Zhu and Xiaoye Qu and Yuchen Fan and Qianshan Wei and Rui Ye and Li Kang and Yiran Qin and Zhiqiang Kou and Daizong Liu and Qi Li and Ning Ding and Siheng Chen and Jing Shao},
      year={2026},
      eprint={2601.14192},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2601.14192}, 
}