mlsanityData to 7 Oct 2026

TroVE: Inducing Verifiable and Efficient Toolboxes for Solving Programmatic Tasks

arXiv 2401.12869 ↗First posted 23 Jan 2024cs.AIcited by 70

Not on the map. Why?

Cited by 39 in LLM agents and tool use, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Language models (LMs) can solve tasks such as answering questions about tables or images by writing programs. However, using primitive functions often leads to verbose and error-prone programs, and higher-level functions require expert design. To enable better solutions without human labor, we ask code LMs to curate reusable high-level functions, and use them to write solutions. We present TROVE, a training-free method of inducing a verifiable and efficient toolbox of functions, by generating via using, growing, and periodically trimming the toolbox. On 11 datasets from math, table question answering, and image reasoning tasks, TROVE consistently yields simpler solutions with higher accuracy than baselines using CODELLAMA and previous methods using GPT, while using 79-98% smaller toolboxes. TROVE further enables 31% faster and 13% more accurate human verification than baselines. With the same pipeline, it creates diverse functions for varied tasks and datasets, providing insights into their individual characteristics.

Key facts. Cited by 70 papers in the map (data through 7 Oct 2026). Builds on 25 of its 33 references that are in the map. First posted on arXiv 23 Jan 2024.

Builds on

25 of 33 references.

References · 25

  1. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  2. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  3. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
  4. 2022Competition-Level Code Generation with AlphaCodecited by 1,427
  5. 2022Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Taskscited by 1,057
  6. 2023ViperGPT: Visual Inference via Python Execution for Reasoningcited by 591
  7. 2023MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Actioncited by 511
  8. 2015Compositional Semantic Parsing on Semi-Structured Tablescited by 509
  9. 2022Visual Programming: Compositional visual reasoning without trainingcited by 507
  10. 2022Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoningcited by 347
  11. 2023Large Language Models as Tool Makerscited by 293
  12. 2023LEVER: Learning to Verify Language-to-Code Generation with Executioncited by 228
13 more
  1. 2022Binding Language Models in Symbolic Languagescited by 224
  2. 2020DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learningcited by 159
  3. 2022Natural Language to Code Translation with Executioncited by 122
  4. 2023CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsetscited by 122
  5. 2021HiTab: A Hierarchical Table Dataset for Question Answering and Natural Language Generationcited by 117
  6. 2023LEGO-Prover: Neural Theorem Proving with Growing Librariescited by 108
  7. 2023AssistGPT: A General Multi-modal Assistant that can Plan, Execute, Inspect, and Learncited by 97
  8. 2023CLIN: A Continually Learning Language Agent for Rapid Task Adaptation and Generalizationcited by 88
  9. 2022Top-Down Synthesis for Library Learningcited by 57
  10. 2023Modular Visual Question Answering via Code Generationcited by 47
  11. 2023LILO: Learning Interpretable Libraries by Compressing and Documenting Codecited by 46
  12. 2021Leveraging Language to Learn Program Abstractions and Search Heuristicscited by 44
  13. 2019Program Synthesis and Semantic Parsing with Learned Code Idiomscited by 34

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 11; 2025: 24; 2026: 35.

70 papers.

Sort:

By area:

  1. 2024BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructionscited by 527
  2. 2024Agent Workflow Memorycited by 342
  3. 2024Tool Learning with Large Language Models: A Surveycited by 221
  4. 2025SkillWeaver: Web Agents can Self-Improve by Discovering and Honing Skillscited by 174
  5. 2025Alita: Generalist Agent Enabling Scalable Agentic Reasoning with Minimal Predefinition and Maximal Self-Evolutioncited by 114
  6. 2025Inducing Programmatic Skills for Agentic Taskscited by 103
  7. 2025How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behaviorcited by 95
  8. 2024SciAgent: Tool-augmented Language Models for Scientific Reasoningcited by 69
  9. 2025SE-Agent: Self-Evolution Trajectory Optimization in Multi-Step Reasoning with LLM-Based Agentscited by 66
  10. 2025AutoPresent: Designing Structured Visuals from Scratchcited by 63
  11. 2024HyperAgent: Generalist Software Engineering Agents to Solve Coding Tasks at Scalecited by 61
  12. 2024What Are Tools Anyway? A Survey from the Language Model Perspectivecited by 53
38 more
  1. 2025SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarioscited by 49
  2. 2026A Survey of Agent Memory in the Second Half: Towards Self-Evolving and Long-Horizon Agentscited by 40
  3. 2024Middleware for LLMs: Tools Are Instrumental for Language Agents in Complex Environmentscited by 39
  4. 2025Visual Large Language Models for Generalized and Specialized Applicationscited by 35
  5. 2024ReGAL: Refactoring Programs to Discover Generalizable Abstractionscited by 32
  6. 2025UItron: Foundational GUI Agent with Advanced Perception and Planningcited by 22
  7. 2025RepoMaster: Autonomous Exploration and Understanding of GitHub Repositories for Complex Task Solvingcited by 19
  8. 2025How Do AI Agents Do Human Work? Comparing AI and Human Workflows Across Diverse Occupationscited by 17
  9. 2026A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applicationscited by 17
  10. 2025ToolMem: Enhancing Multimodal Agents with Learnable Tool Capability Memorycited by 14
  11. 2024PoTable: Towards Systematic Thinking via Plan-then-Execute Stage Reasoning on Tablescited by 13
  12. 2026SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learningcited by 13
  13. 2025From Reproduction to Replication: Evaluating Research Agents with Progressive Code Maskingcited by 12
  14. 2026Toward Efficient Agents: Memory, Tool learning, and Planningcited by 11
  15. 2025Benchmarking Failures in Tool-Augmented Language Modelscited by 9
  16. 2025Table-r1: Self-supervised and Reinforcement Learning for Program-based Table Reasoning in Small Language Modelscited by 7
  17. 2026User as Code: Executable Memory for Personalized Agentscited by 7
  18. 2025ToolLibGen: Scalable Automatic Tool Creation and Aggregation for LLM Reasoningcited by 5
  19. 2025Automated Cloud Infrastructure-as-Code Reconciliation with AI Agentscited by 5
  20. 2024PropTest: Automatic Property Testing for Improved Visual Programmingcited by 4
  21. 2025Refactoring Codebases through Library Designcited by 4
  22. 2026CLEANER: Self-Purified Trajectories Boost Agentic Reinforcement Learningcited by 4
  23. 2026PreAct: Computer-Using Agents that Get Faster on Repeated Taskscited by 4
  24. 2024Library Learning Doesn't: The Curious Case of the Single-Use "Library"cited by 3
  25. 2025GATE: Graph-based Adaptive Tool Evolution Across Diverse Taskscited by 3
  26. 2025LLM Library Learning Fails: A LEGO-Prover Case Studycited by 3
  27. 2025Divide, Optimize, Merge: Fine-Grained LLM Agent Optimization at Scalecited by 3
  28. 2026SKILL.nb: Selective Formalization and Gated Execution for Durable Agent Workflowscited by 3
  29. 2026Humans are Missing from AI Coding Agent Researchcited by 3
  30. 2025A Compute-Matched Re-Evaluation of TroVE on MATHcited by 2
  31. 2025Transductive Visual Programming: Evolving Tool Libraries from Experience for Spatial Reasoningcited by 2
  32. 2026Agentic Tool Use in Large Language Modelscited by 2
  33. 2026Online Skill Learning for Web Agents via State-Grounded Dynamic Retrievalcited by 2
  34. 2026VASO: Formally Verifiable Self-Evolving Skills for Physical AI Agentscited by 2
  35. 2026Tool-Making and Self-Evolving LLM Agents in Low-Latency Systemscited by 2
  36. 2026PhoenixRepair: Rethinking Repair Strategy Exploration in Software Agentscited by 2
  37. 2026AllocBench: Measuring Online Tool Allocation Capability in LLM Agentscited by 2
  38. 2026A Frozen 12B Beats Frontier Models on Verified Work: 100% Accuracy, 0 Tokens, Bit-Exact, Forevercited by 2

50 most cited of 70 All 70 (JSON)