# ReGAL: Refactoring Programs to Discover Generalizable Abstractions

arXiv: https://arxiv.org/abs/2401.16467 · HTML: https://mlsanity.com/paper/2401.16467/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Builds on 41 of its 71 references that are in the map.
- First posted on arXiv 29 Jan 2024.

## Builds on (most cited first, 30 of 41)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (2023)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [AgentBench: Evaluating LLMs as Agents](https://mlsanity.com/paper/2308.03688/) (2023)
- [What Makes Good In-Context Examples for GPT-3?](https://mlsanity.com/paper/2101.06804/) (2021)
- [Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks](https://mlsanity.com/paper/2211.12588/) (2022)
- [Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents](https://mlsanity.com/paper/2201.07207/) (2022)
- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models](https://mlsanity.com/paper/2307.05973/) (2023)
- [ViperGPT: Visual Inference via Python Execution for Reasoning](https://mlsanity.com/paper/2303.08128/) (2023)
- [Decomposed Prompting: A Modular Approach for Solving Complex Tasks](https://mlsanity.com/paper/2210.02406/) (2022)
- [Identifying the Risks of LM Agents with an LM-Emulated Sandbox](https://mlsanity.com/paper/2309.15817/) (2023)
- [Visual Programming: Compositional visual reasoning without training](https://mlsanity.com/paper/2211.11559/) (2022)
- [Faith and Fate: Limits of Transformers on Compositionality](https://mlsanity.com/paper/2305.18654/) (2023)
- [Augmented Language Models: a Survey](https://mlsanity.com/paper/2302.07842/) (2023)
- [Neural Module Networks](https://mlsanity.com/paper/1511.02799/) (2015)
- [Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models](https://mlsanity.com/paper/2304.09842/) (2023)
- [Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning](https://mlsanity.com/paper/2209.14610/) (2022)
- [TextWorld: A Learning Environment for Text-based Games](https://mlsanity.com/paper/1806.11532/) (2018)
- [Large Language Models as Tool Makers](https://mlsanity.com/paper/2305.17126/) (2023)
- [Faithful Chain-of-Thought Reasoning](https://mlsanity.com/paper/2301.13379/) (2023)
- [ADaPT: As-Needed Decomposition and Planning with Language Models](https://mlsanity.com/paper/2311.05772/) (2023)
- [CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsets](https://mlsanity.com/paper/2309.17428/) (2023)
- [Large Language Models as Analogical Reasoners](https://mlsanity.com/paper/2310.01714/) (2023)
- [Semantic Parsing for Task Oriented Dialog using Hierarchical Representations](https://mlsanity.com/paper/1810.07942/) (2018)

## Built on by (most cited first, 30 of 36)

- [When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs](https://mlsanity.com/paper/2406.01297/) (2024)
- [A Survey on Large Language Model-Based Game Agents](https://mlsanity.com/paper/2404.02039/) (2024)
- [What Are Tools Anyway? A Survey from the Language Model Perspective](https://mlsanity.com/paper/2403.15452/) (2024)
- [Is Programming by Example solved by LLMs?](https://mlsanity.com/paper/2406.08316/) (2024)
- [Learning to Refine with Fine-Grained Natural Language Feedback](https://mlsanity.com/paper/2407.02397/) (2024)
- [LLMs Can Generate a Better Answer by Aggregating Their Own Responses](https://mlsanity.com/paper/2503.04104/) (2025)
- [Empowering Large Language Model Agents through Action Learning](https://mlsanity.com/paper/2402.15809/) (2024)
- [The Fusion of Large Language Models and Formal Methods for Trustworthy AI Agents: A Roadmap](https://mlsanity.com/paper/2412.06512/) (2024)
- [Pix2Code: Learning to Compose Neural Visual Concepts as Programs](https://mlsanity.com/paper/2402.08280/) (2024)
- [Generator-Assistant Stepwise Rollback Framework for Large Language Model Agent](https://mlsanity.com/paper/2503.02519/) (2025)
- [One Life to Learn: Inferring Symbolic World Models for Stochastic Environments from Unguided Exploration](https://mlsanity.com/paper/2510.12088/) (2025)
- [Enhancing Agent Learning through World Dynamics Modeling](https://mlsanity.com/paper/2407.17695/) (2024)
- [Evolving Programmatic Skill Networks](https://mlsanity.com/paper/2601.03509/) (2026)
- [ToolLibGen: Scalable Automatic Tool Creation and Aggregation for LLM Reasoning](https://mlsanity.com/paper/2510.07768/) (2025)
- [Experience Transfer for Multimodal LLM Agents in Minecraft Game](https://mlsanity.com/paper/2604.05533/) (2026)
- [Action abstractions for amortized sampling](https://mlsanity.com/paper/2410.15184/) (2024)
- [Library Learning Doesn't: The Curious Case of the Single-Use "Library"](https://mlsanity.com/paper/2410.20274/) (2024)
- [Refactoring Codebases through Library Design](https://mlsanity.com/paper/2506.11058/) (2025)
- [GATE: Graph-based Adaptive Tool Evolution Across Diverse Tasks](https://mlsanity.com/paper/2502.14848/) (2025)
- [Test-Time Learning with an Evolving Library](https://mlsanity.com/paper/2605.14477/) (2026)
- [A Compute-Matched Re-Evaluation of TroVE on MATH](https://mlsanity.com/paper/2507.22069/) (2025)
- [ReaComp: Compiling LLM Reasoning into Symbolic Solvers for Efficient Program Synthesis](https://mlsanity.com/paper/2605.05485/) (2026)
- [AllocBench: Measuring Online Tool Allocation Capability in LLM Agents](https://mlsanity.com/paper/2607.23332/) (2026)
- [BIM Information Extraction Through LLM-based Adaptive Exploration](https://mlsanity.com/paper/2605.01698/) (2026)
- [Better, Faster, Stronger: Programmatic Skill Learning Best Reduces Agent Cost](https://mlsanity.com/paper/2608.11338/) (2026)
- [CoCoDA: Co-evolving Compositional DAG for Tool-Augmented Agents](https://mlsanity.com/paper/2605.08399/) (2026)
- [PYTHALAB-MERA: Validation-Grounded Memory, Retrieval, and Acceptance Control for Frozen-LLM Coding Agents](https://mlsanity.com/paper/2605.08468/) (2026)
- [XSearch: Explainable Code Search via Concept-to-Code Alignment](https://mlsanity.com/paper/2605.16046/) (2026)
- [PACE: Primitive-Aware Code Evolution for Automated Algorithm Design](https://mlsanity.com/paper/2608.07395/) (2026)
- [Credo: Reusable Declarative Primitives for Agentic Workflows](https://mlsanity.com/paper/2608.27790/) (2026)

## Abstract (arXiv)

While large language models (LLMs) are increasingly being used for program synthesis, they lack the global view needed to develop useful abstractions; they generally predict programs one at a time, often repeating the same functionality. Generating redundant code from scratch is both inefficient and error-prone. To address this, we propose Refactoring for Generalizable Abstraction Learning (ReGAL), a gradient-free method for learning a library of reusable functions via code refactorization, i.e., restructuring code without changing its execution output. ReGAL learns from a small set of existing programs, iteratively verifying and refining its abstractions via execution. We find that the shared function libraries discovered by ReGAL make programs easier to predict across diverse domains. On five datasets -- LOGO graphics generation, Date reasoning, TextCraft (a Minecraft-based text-game) MATH, and TabMWP -- both open-source and proprietary LLMs improve in accuracy when predicting programs with ReGAL functions. For CodeLlama-13B, ReGAL results in absolute accuracy increases of 11.5% on LOGO, 26.1% on date understanding, and 8.1% on TextCraft, outperforming GPT-3.5 in two of three domains. Our analysis reveals ReGAL's abstractions encapsulate frequently-used subroutines as well as environment dynamics.