mlsanityData to 14 Oct 2026

MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback

arXiv 2309.10691 ↗First posted 19 Sep 2023cs.CL cs.AI cs.LGcited by 297

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

also in 1 more
Abstract (arXiv)

To solve complex tasks, large language models (LLMs) often require multiple rounds of interactions with the user, sometimes assisted by external tools. However, current evaluation protocols often emphasize benchmark performance with single-turn exchanges, neglecting the nuanced interactions among the user, LLMs, and external tools, while also underestimating the importance of natural language feedback from users. These oversights contribute to discrepancies between research benchmark evaluations and real-world use cases. We introduce MINT, a benchmark that evaluates LLMs' ability to solve tasks with multi-turn interactions by (1) using tools and (2) leveraging natural language feedback. To ensure reproducibility, we provide an evaluation framework where LLMs can access tools by executing Python code and receive users' natural language feedback simulated by GPT-4. We repurpose a diverse set of established evaluation datasets focusing on reasoning, coding, and decision-making and carefully curate them into a compact subset for efficient evaluation. Our analysis of 20 open- and closed-source LLMs offers intriguing findings. (a) LLMs generally benefit from tools and language feedback, with performance gains (absolute, same below) of 1-8% for each turn of tool use and 2-17% with natural language feedback. (b) Better single-turn performance does not guarantee better multi-turn performance. (c) Surprisingly, on the LLMs evaluated, supervised instruction-finetuning (SIFT) and reinforcement learning from human feedback (RLHF) generally hurt multi-turn capabilities. We expect MINT can help measure progress and incentivize research in improving LLMs' capabilities in multi-turn interactions, especially for open-source communities where multi-turn human evaluation can be less accessible compared to commercial LLMs with a larger user base.

Key facts. Cited by 297 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 58 of its 77 references that are in the map. First posted on arXiv 19 Sep 2023.

Builds on

58 of 77 references.

Closest ancestors · 8

Same direction, newest first

  1. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs31 Jul
  2. 2023ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Cases8 Jun
  3. 2023Gorilla: Large Language Model Connected with Massive APIs24 May
  4. 2023ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings19 May
  5. 2023Tool Learning with Foundation Models17 Apr
  6. 2023ART: Automatic multi-step reasoning and tool-use for large language models16 Mar
  7. 2023Toolformer: Language Models Can Teach Themselves to Use Tools9 Feb
  8. 2022TALM: Tool Augmented Language Models24 May

Other references · 50

  1. 2022Training language models to follow instructions with human feedbackcited by 15,971
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  3. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  5. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  6. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  7. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  8. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  9. 2021Finetuned Language Models Are Zero-Shot Learnerscited by 3,780
  10. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  11. 2021Program Synthesis with Large Language Modelscited by 3,637
  12. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
38 more
  1. 2023Code Llama: Open Foundation Models for Codecited by 2,139
  2. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 2,006
  3. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,745
  4. 2023AgentBench: Evaluating LLMs as Agentscited by 1,267
  5. 2023Mind2Web: Towards a Generalist Agent for the Webcited by 1,242
  6. 2020ALFWorld: Aligning Text and Embodied Environments for Interactive Learningcited by 1,205
  7. 2022Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Taskscited by 1,076
  8. 2022CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesiscited by 885
  9. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 783
  10. 2022PAL: Program-aided Language Modelscited by 652
  11. 2023LeanDojo: Theorem Proving with Retrieval-Augmented Language Modelscited by 427
  12. 2023Augmented Language Models: a Surveycited by 403
  13. 2022Self-critiquing models for assisting human evaluatorscited by 380
  14. 2023Chameleon: Plug-and-Play Compositional Reasoning with Large Language Modelscited by 356
  15. 2023Check Your Facts and Try Again: Improving Large Language Models with External Knowledge and Automated Feedbackcited by 353
  16. 2018TextWorld: A Learning Environment for Text-based Gamescited by 310
  17. 2023AudioGPT: Understanding and Generating Speech, Music, Sound, and Talking Headcited by 265
  18. 2022CoAuthor: Designing a Human-AI Collaborative Writing Dataset for Exploring Language Model Capabilitiescited by 249
  19. 2023InterCode: Standardizing and Benchmarking Interactive Coding with Execution Feedbackcited by 238
  20. 2023Improving Language Model Negotiation with Self-Play and In-Context Learning from AI Feedbackcited by 207
  21. 2022BlenderBot 3: a deployed conversational agent that continually learns to responsibly engagecited by 198
  22. 2023TheoremQA: A Theorem-driven Question Answering datasetcited by 192
  23. 2023Fact-Checking Complex Claims with Program-Guided Reasoningcited by 146
  24. 2023WebGLM: Towards An Efficient Web-Enhanced Question Answering System with Human Preferencescited by 100
  25. 2021A Neural Network Solves, Explains, and Generates University Math Problems by Program Synthesis and Few-Shot Learning at Human Levelcited by 96
  26. 2022Evaluating Human-Language Model Interactioncited by 92
  27. 2022PEER: A Collaborative Language Modelcited by 86
  28. 2022Mind's Eye: Grounded Language Model Reasoning through Simulationcited by 84
  29. 2023RewriteLM: An Instruction-Tuned Large Language Model for Text Rewritingcited by 67
  30. 2022POTATO: The Portable Text Annotation Toolcited by 64
  31. 2023HAGRID: A Human-LLM Collaborative Dataset for Generative Information-Seeking with Attributioncited by 56
  32. 2023Bridging the Gap: A Survey on Integrating (Human) Feedback for Natural Language Generationcited by 55
  33. 2020Speak to your Parser: Interactive Text-to-SQL with Natural Language Feedbackcited by 44
  34. 2023LatEval: An Interactive LLMs Evaluation Benchmark with Incomplete Information from Lateral Thinking Puzzlescited by 41
  35. 2019Model-based Interactive Semantic Parsing: A Unified Framework and A Text-to-SQL Case Studycited by 40
  36. 2022Learning to Model Editing Processescited by 30
  37. 2023LeTI: Learning to Generate from Textual Interactionscited by 25
  38. 2023Learning to Simulate Natural Language Feedback for Interactive Semantic Parsingcited by 19

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 20; 2024: 81; 2025: 100; 2026: 96.

297 papers.

Sort:

By area:

  1. 2024OpenHands: An Open Platform for AI Software Developers as Generalist Agentscited by 1,002
  2. 2024Executable Code Actions Elicit Better LLM Agentscited by 564
  3. 2025LLMs Get Lost In Multi-Turn Conversationcited by 429
  4. 2023UltraFeedback: Boosting Language Models with Scaled AI Feedbackcited by 406
  5. 2025RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learningcited by 386
  6. 2024AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agentscited by 339
  7. 2024ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilitiescited by 295
  8. 2024MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialoguescited by 250
  9. 2025LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Trainingcited by 244
  10. 2024Leak, Cheat, Repeat: Data Contamination and Evaluation Malpractices in Closed-Source LLMscited by 229
  11. 2024Tool Learning with Large Language Models: A Surveycited by 223
  12. 2024AI Agents That Mattercited by 219
38 more
  1. 2024From Persona to Personalization: A Survey on Role-Playing Language Agentscited by 212
  2. 2025Survey on Evaluation of LLM-based Agentscited by 204
  3. 2024APIGen: Automated Pipeline for Generating Verifiable and Diverse Function-Calling Datasetscited by 198
  4. 2024Trial and Error: Exploration-Based Trajectory Optimization for LLM Agentscited by 197
  5. 2024Advancing LLM Reasoning Generalists with Preference Treescited by 192
  6. 2023Navigate through Enigmatic Labyrinth A Survey of Chain of Thought Reasoning: Advances, Frontiers and Futurecited by 184
  7. 2024SciCode: A Research Coding Benchmark Curated by Scientistscited by 178
  8. 2023Removing RLHF Protections in GPT-4 via Fine-Tuningcited by 171
  9. 2024Cost-Efficient Large Language Model Serving for Multi-turn Conversations with CachedAttentioncited by 163
  10. 2023OpenAgents: An Open Platform for Language Agents in the Wildcited by 137
  11. 2024StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Modelscited by 132
  12. 2023CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsetscited by 129
  13. 2024GTBench: Uncovering the Strategic Reasoning Limitations of LLMs via Game-Theoretic Evaluationscited by 120
  14. 2025SWEET-RL: Training Multi-Turn LLM Agents on Collaborative Reasoning Taskscited by 118
  15. 2024xLAM: A Family of Large Action Models to Empower AI Agent Systemscited by 111
  16. 2024LLM Agents can Autonomously Hack Websitescited by 109
  17. 2024HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Modelcited by 106
  18. 2023DRESS: Instructing Large Vision-Language Models to Align and Interact with Humans via Natural Language Feedbackcited by 103
  19. 2024LLaMA Pro: Progressive LLaMA with Block Expansioncited by 101
  20. 2024AIOS: LLM Agent Operating Systemcited by 97
  21. 2024Internet of Agents: Weaving a Web of Heterogeneous Agents for Collaborative Intelligencecited by 86
  22. 2024MT-Eval: A Multi-Turn Capabilities Evaluation Benchmark for Large Language Modelscited by 81
  23. 2024If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agentscited by 79
  24. 2024EHRAgent: Code Empowers Large Language Models for Few-shot Complex Tabular Reasoning on Electronic Health Recordscited by 78
  25. 2025UserBench: An Interactive Gym Environment for User-Centric Agentscited by 76
  26. 2023Lemur: Harmonizing Natural Language and Code for Language Agentscited by 72
  27. 2024SciAgent: Tool-augmented Language Models for Scientific Reasoningcited by 70
  28. 2024Preble: Efficient Distributed Prompt Scheduling for LLM Servingcited by 70
  29. 2024OfficeBench: Benchmarking Language Agents across Multiple Applications for Office Automationcited by 70
  30. 2023AlignBench: Benchmarking Chinese Alignment of Large Language Modelscited by 68
  31. 2024StateFlow: Enhancing LLM Task-Solving through State-Driven Workflowscited by 65
  32. 2026Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agentscited by 64
  33. 2024Aligning LLMs with Individual Preferences via Interactioncited by 63
  34. 2023Language Models, Agent Models, and World Models: The LAW for Machine Reasoning and Planningcited by 60
  35. 2024Agents in Software Engineering: Survey, Landscape, and Visioncited by 60
  36. 2024SCBench: A KV Cache-Centric Analysis of Long-Context Methodscited by 60
  37. 2024Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Trainingcited by 57
  38. 2025Evaluating LLM-based Agents for Multi-Turn Conversations: A Surveycited by 57

50 most cited of 297 All 297 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Beyond Ideal Instruction: A Comprehensive Framework for Evaluating LLMs in Realistic Interactions
  2. 2025ACEBench: Who Wins the Match Point in Tool Usage?
  3. 2024Can Large Language Models be Trusted for Evaluation? Scalable Meta-Evaluation of LLMs as Evaluators via Agent Debate
  4. 2025CLEAR: Error Analysis via LLM-as-a-Judge Made Easy
  5. 2025ToolHop: A Query-Driven Benchmark for Evaluating Large Language Models in Multi-Hop Tool Use
  6. 2025LMR-BENCH: Evaluating LLM Agent's Ability on Reproducing Language Modeling Research
  7. 2025CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks
  8. 2025Cross-Task Benchmarking and Evaluation of General-Purpose and Code-Specific Large Language Models ↗