mlsanityData to 7 Oct 2026

The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration

arXiv 2603.22862 ↗First posted 24 Mar 2026cs.SE cs.CLcited by 11

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

Abstract (arXiv)

Tool use enables large language models (LLMs) to access external information, invoke software systems, and act in digital environments beyond what can be solved from model parameters alone. Early research mainly studied whether a model could select and execute a correct single tool call. As agent systems evolve, however, the central problem has shifted from isolated invocation to multi-tool orchestration over long trajectories with intermediate state, execution feedback, changing environments, and practical constraints such as safety, cost, and verifiability. We comprehensively review recent progress in multi-tool LLM agents and analyzes the state of the art in this rapidly developing area. First, we unify task formulations and distinguish single-call tool use from long-horizon orchestration. Then, we organize the literature around six core dimensions: inference-time planning and execution, training and trajectory construction, safety and control, efficiency under resource constraints, capability completeness in open environments, and benchmark design and evaluation. We further summarize representative applications in software engineering, enterprise workflows, graphical user interfaces, and mobile systems. Finally, we discuss major challenges and outline future directions for building reliable, scalable, and verifiable multi-tool agents.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Placed in Tool learning and function calling, one of 16 directions in LLM agents and tool use. Builds on 187 of its 226 references that are in the map. First posted on arXiv 24 Mar 2026.

Builds on

187 of 226 references.

Closest ancestors · 45

Same direction, newest first

  1. 2025M^3-Bench: Multi-Modal, Multi-Hop, Multi-Threaded Tool-Using MLLM Agent Benchmark21 Nov
  2. 2025AutoTool: Efficient Tool Selection for Large Language Model Agents18 Nov
  3. 2025Tool-to-Agent Retrieval: Bridging Tools and Agents for Scalable LLM Multi-Agent Systems3 Nov
  4. 2025ToolDreamer: Instilling LLM Reasoning Into Tool Retrievers22 Oct
  5. 2025TRAJECT-Bench:A Trajectory-Aware Benchmark for Evaluating Agentic Tool Use6 Oct
  6. 2025Towards Reliable Benchmarking: A Contamination Free, Controllable Evaluation Framework for Multi-step LLM Function Calling30 Sep
  7. 2025Failure Makes the Agent Stronger: Enhancing Accuracy through Structured Reflection for Reliable Tool Interactions23 Sep
  8. 2025MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Servers28 Aug
  9. 2025Butterfly Effects in Toolchains: A Comprehensive Analysis of Failed Parameter Filling in LLM Tool-Agent Systems21 Jul
  10. 2025DICE-BENCH: Evaluating the Tool-Use Capabilities of Large Language Models in Multi-Round, Multi-Party Dialogues28 Jun
  11. 2025CRITICTOOL: Evaluating Self-Critique Capabilities of Large Language Models in Tool-Calling Error Scenarios11 Jun
  12. 2025MCP-Zero: Active Tool Discovery for Autonomous LLM Agents1 Jun
33 more
  1. 2025ToolHaystack: Stress-Testing Tool-Augmented Language Models in Realistic Long-Term Interactions29 May
  2. 2025Rethinking Stateful Tool Use in Multi-Turn Dialogues: Benchmarks and Challenges19 May
  3. 2025ToolACE-DEV: Self-Improving Tool Learning via Decomposition and EVolution12 May
  4. 2025ToolACE-R: Model-aware Iterative Training and Adaptive Refinement for Tool Learning2 Apr
  5. 2025StableToolBench-MirrorAPI: Modeling Tool Environments as Mirrors of 7,000+ Real-World APIs26 Mar
  6. 2025Chain-of-Tools: Utilizing Massive Unseen Tools in the CoT Reasoning of Frozen Language Models21 Mar
  7. 2025PLAY2PROMPT: Zero-shot Tool Instruction Optimization for LLM Agents via Tool Play18 Mar
  8. 2025Benchmarking Failures in Tool-Augmented Language Models18 Mar
  9. 2025AskToAct: Enhancing LLMs Tool Use via Self-Correcting Clarification3 Mar
  10. 2025Self-Training Large Language Models for Tool-Use Without Demonstrations9 Feb
  11. 2025Divide-Then-Aggregate: An Efficient Tool Learning Method via Parallel Tool Invocation21 Jan
  12. 2025ComplexFuncBench: Exploring Multi-Step and Constrained Function Calling under Long-Context Scenario17 Jan
  13. 2025ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool Use5 Jan
  14. 2024Facilitating Multi-turn Function Calling for LLMs via Compositional Instruction Tuning16 Oct
  15. 2024From Exploration to Mastery: Enabling LLMs to Master Tools via Self-Driven Interactions10 Oct
  16. 2024Hammer: Robust Function-Calling for On-Device Language Models via Function Masking6 Oct
  17. 2024NESTFUL: A Benchmark for Evaluating LLMs on Nested Sequences of API Calls4 Sep
  18. 2024Granite-Function Calling Model: Introducing Function Calling Abilities via Multi-task Learning of Granular Tasks27 Jun
  19. 2024APIGen: Automated Pipeline for Generating Verifiable and Diverse Function-Calling Datasets26 Jun
  20. 2024Seal-Tools: Self-Instruct Tool Learning Dataset for Agent Tuning and Detailed Benchmark14 May
  21. 2024ToolSword: Unveiling Safety Issues of Large Language Models in Tool Learning Across Three Stages16 Feb
  22. 2024AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Calls6 Feb
  23. 2024Efficient Tool Use with Chain-of-Abstraction Reasoning30 Jan
  24. 2024RoTBench: A Multi-Level Benchmark for Evaluating the Robustness of Large Language Models in Tool Learning16 Jan
  25. 2023MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use4 Oct
  26. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs31 Jul
  27. 2023GPT4Tools: Teaching Large Language Model to Use Tools via Self-instruction30 May
  28. 2023Large Language Models as Tool Makers26 May
  29. 2023Gorilla: Large Language Model Connected with Massive APIs24 May
  30. 2023Making Language Models Better Tool Learners with Execution Feedback22 May
  31. 2023ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings19 May
  32. 2023API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs14 Apr
  33. 2023Toolformer: Language Models Can Teach Themselves to Use Tools9 Feb

Other references · 142

  1. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  2. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  3. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,383
  4. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
  5. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,615
  6. 2023Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injectioncited by 1,339
  7. 2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domainscited by 1,280
  8. 2023HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Facecited by 1,173
  9. 2024OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environmentscited by 1,134
  10. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,000
  11. 2023ExpeL: LLM Agents Are Experiential Learnerscited by 814
  12. 2023DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelinescited by 803
38 more
  1. 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 791
  2. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 751
  3. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 633
  4. 2023Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Modelscited by 551
  5. 2024GPT-4V(ision) is a Generalist Web Agent, if Groundedcited by 547
  6. 2024InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agentscited by 476
  7. 2023AppAgent: Multimodal Agents as Smartphone Userscited by 407
  8. 2024Agentless: Demystifying LLM-based Software Engineering Agentscited by 400
  9. 2024Navigating the Digital World as Humans Do: Universal Visual Grounding for GUI Agentscited by 373
  10. 2024AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agentscited by 359
  11. 2025RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learningcited by 357
  12. 2024WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?cited by 348
  13. 2023RepoBench: Benchmarking Repository-Level Code Auto-Completion Systemscited by 346
  14. 2024ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilitiescited by 272
  15. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactionscited by 236
  16. 2024Spider 2.0: Evaluating Language Models on Real-World Enterprise Text-to-SQL Workflowscited by 223
  17. 2024Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scalecited by 212
  18. 2025AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agentscited by 196
  19. 2025Plan-and-Act: Improving Planning of Agents for Long-Horizon Taskscited by 190
  20. 2024WebLINX: Real-World Website Navigation with Multi-Turn Dialoguecited by 177
  21. 2023ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Modelscited by 174
  22. 2023ADaPT: As-Needed Decomposition and Planning with Language Modelscited by 172
  23. 2023An LLM Compiler for Parallel Function Callingcited by 170
  24. 2023SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Taskscited by 166
  25. 2025MIRIX: Multi-Agent Memory System for LLM-Based Agentscited by 160
  26. 2022MRKL Systems: A modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoningcited by 153
  27. 2025Evaluation and Benchmarking of LLM Agents: A Surveycited by 145
  28. 2023TaskBench: Benchmarking Large Language Models for Task Automationcited by 138
  29. 2024The Emerged Security and Privacy of LLM Agent: A Survey with Case Studiescited by 127
  30. 2025Memory Injection Attacks on LLM Agents via Query-Only Interactioncited by 127
  31. 2024ST-WebAgentBench: A Benchmark for Evaluating Safety and Trustworthiness in Web Agentscited by 126
  32. 2025ACON: Optimizing Context Compression for Long-horizon LLM Agentscited by 126
  33. 2025Memento: Fine-tuning LLM Agents without Fine-tuning LLMscited by 119
  34. 2024AgentOccam: A Simple Yet Strong Baseline for LLM-Based Web Agentscited by 110
  35. 2024AndroidLab: Training and Systematic Benchmarking of Android Autonomous Agentscited by 105
  36. 2023Skeleton-of-Thought: Prompting LLMs for Efficient Parallel Generationcited by 103
  37. 2024ScreenAgent: A Vision Language Model-driven Computer Control Agentcited by 101
  38. 2024Small LLMs Are Weak Tool Learners: A Multi-LLM Agentcited by 92

50 most cited of 142 All 142 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 11.

11 papers.

  1. 2026From Skill Text to Skill Structure: The Scheduling-Structural-Logical Representation for Agent Skillscited by 15
  2. 2026Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memorycited by 3
  3. 2026FinTrace: Holistic Trajectory-Level Evaluation of LLM Tool Calling for Long-Horizon Financial Taskscited by 2
  4. 2026HarnessAPI: A Skill-First Framework for Unified Streaming APIs and MCP Toolscited by 1
  5. 2026Accuracy Is Speed: Towards Long-Context-Aware Routing for Distributed LLM Servingcited by 0
  6. 2026OpenClaw and Ollama in Agentic AI: Toward Fully Autonomous and Scalable AI Agent Systemscited by 0
  7. 2026SPA: Securing Persistent LLM Agents Across Queries with Plan-First Information-Flow Controlcited by 0
  8. 2026Speculative Macro Commit for Faster Tool-Using Agentscited by 0
  9. 2026Compositional SVG Generation via VLM-Driven Hierarchical Semantic Parsingcited by 0
  10. 2026Symbolic Temporal Supervision of LLM Agents Using Contractscited by 0
  11. 2026Learning What to Activate: Combinatorial Capability Allocation for Long-Horizon Multimodal Agentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Utility-Guided Agent Orchestration for Efficient LLM Tool Use
  2. 2026How Consistent Are LLM Agents? Measuring Behavioral Reproducibility in Multi-Step Tool-Calling Pipelines
  3. 2025Tool Preferences in Agentic LLMs are Unreliable
  4. 2026DOCSCHISEL: Adaptive Tool Documentation Optimization Framework for LLM Agents
  5. 2025A Comprehensive Survey on Benchmarks and Solutions in Software Engineering of LLM-Empowered Agentic System
  6. 2024Large Language Model-Based Agents for Software Engineering: A Survey
  7. 2025Multi-Mission Tool Bench: Assessing the Robustness of LLM based Agents through Related and Dynamic Missions
  8. 2025LoCoBench-Agent: An Interactive Benchmark for LLM Agents in Long-Context Software Engineering