# LILO: Learning Interpretable Libraries by Compressing and Documenting Code

arXiv: https://arxiv.org/abs/2310.19791 · HTML: https://mlsanity.com/paper/2310.19791/

## Key facts

- Cited by 49 papers in the map (data through 14 Oct 2026).
- Builds on 44 of its 88 references that are in the map.
- First posted on arXiv 30 Oct 2023.

## Builds on (most cited first, 30 of 44)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Competition-Level Code Generation with AlphaCode](https://mlsanity.com/paper/2203.07814/) (2022)
- [CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning](https://mlsanity.com/paper/1612.06890/) (2016)
- [What Makes Good In-Context Examples for GPT-3?](https://mlsanity.com/paper/2101.06804/) (2021)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [Large Language Models Are Human-Level Prompt Engineers](https://mlsanity.com/paper/2211.01910/) (2022)
- [Measuring Coding Challenge Competence With APPS](https://mlsanity.com/paper/2105.09938/) (2021)
- [CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis](https://mlsanity.com/paper/2203.13474/) (2022)
- [InCoder: A Generative Model for Code Infilling and Synthesis](https://mlsanity.com/paper/2204.05999/) (2022)
- [Neural Module Networks](https://mlsanity.com/paper/1511.02799/) (2015)
- [Chain-of-Verification Reduces Hallucination in Large Language Models](https://mlsanity.com/paper/2309.11495/) (2023)
- [Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding](https://mlsanity.com/paper/1810.02338/) (2018)
- [Memorizing Transformers](https://mlsanity.com/paper/2203.08913/) (2022)
- [Synchromesh: Reliable code generation from pre-trained language models](https://mlsanity.com/paper/2201.11227/) (2022)
- [DeepCoder: Learning to Write Programs](https://mlsanity.com/paper/1611.01989/) (2016)
- [RobustFill: Neural Program Learning under Noisy I/O](https://mlsanity.com/paper/1703.07469/) (2017)
- [Learning to Reason: End-to-End Module Networks for Visual Question Answering](https://mlsanity.com/paper/1704.05526/) (2017)
- [Skill Induction and Planning with Latent Language](https://mlsanity.com/paper/2110.01517/) (2021)
- [Natural Language to Code Generation in Interactive Data Science Notebooks](https://mlsanity.com/paper/2212.09248/) (2022)
- [Neuro-Symbolic Program Synthesis](https://mlsanity.com/paper/1611.01855/) (2016)
- [Language Models Can Teach Themselves to Program Better](https://mlsanity.com/paper/2207.14502/) (2022)
- [Learning with Latent Language](https://mlsanity.com/paper/1711.00482/) (2017)
- [Learning to Infer Graphics Programs from Hand-Drawn Images](https://mlsanity.com/paper/1707.09627/) (2017)
- [DOBF: A Deobfuscation Pre-Training Objective for Programming Languages](https://mlsanity.com/paper/2102.07492/) (2021)

## Built on by (most cited first, 30 of 49)

- [Agent Workflow Memory](https://mlsanity.com/paper/2409.07429/) (2024)
- [Inducing Programmatic Skills for Agentic Tasks](https://mlsanity.com/paper/2504.06821/) (2025)
- [WorldCoder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environment](https://mlsanity.com/paper/2402.12275/) (2024)
- [Towards Guaranteed Safe AI: A Framework for Ensuring Robust and Reliable AI Systems](https://mlsanity.com/paper/2405.06624/) (2024)
- [Symbolic Regression with a Learned Concept Library](https://mlsanity.com/paper/2409.09359/) (2024)
- [TroVE: Inducing Verifiable and Efficient Toolboxes for Solving Programmatic Tasks](https://mlsanity.com/paper/2401.12869/) (2024)
- [What Are Tools Anyway? A Survey from the Language Model Perspective](https://mlsanity.com/paper/2403.15452/) (2024)
- [ReGAL: Refactoring Programs to Discover Generalizable Abstractions](https://mlsanity.com/paper/2401.16467/) (2024)
- [Pix2Code: Learning to Compose Neural Visual Concepts as Programs](https://mlsanity.com/paper/2402.08280/) (2024)
- [SkillDAG: Self-Evolving Typed Skill Graphs for LLM Skill Selection at Scale](https://mlsanity.com/paper/2606.03056/) (2026)
- [Modeling Open-World Cognition as On-Demand Synthesis of Probabilistic Models](https://mlsanity.com/paper/2507.12547/) (2025)
- [Executable World Models for ARC-AGI-3 in the Era of Coding Agents](https://mlsanity.com/paper/2605.05138/) (2026)
- [Synthesizing Programmatic Reinforcement Learning Policies with Large Language Model Guided Search](https://mlsanity.com/paper/2405.16450/) (2024)
- [Self-Steering Language Models](https://mlsanity.com/paper/2504.07081/) (2025)
- [Loose LIPS Sink Ships: Asking Questions in Battleship with Language-Informed Program Sampling](https://mlsanity.com/paper/2402.19471/) (2024)
- [AI Agents as Universal Task Solvers](https://mlsanity.com/paper/2510.12066/) (2025)
- [ShapeLib: Designing a library of programmatic 3D shape abstractions with Large Language Models](https://mlsanity.com/paper/2502.08884/) (2025)
- [DreamProver: Evolving Transferable Lemma Libraries via a Wake-Sleep Theorem-Proving Agent](https://mlsanity.com/paper/2604.26311/) (2026)
- [Library Learning Doesn't: The Curious Case of the Single-Use "Library"](https://mlsanity.com/paper/2410.20274/) (2024)
- [TurtleBench: A Visual Programming Benchmark in Turtle Geometry](https://mlsanity.com/paper/2411.00264/) (2024)
- [Refactoring Codebases through Library Design](https://mlsanity.com/paper/2506.11058/) (2025)
- [Artificial Intelligence and the Structure of Mathematics](https://mlsanity.com/paper/2604.06107/) (2026)
- [GATE: Graph-based Adaptive Tool Evolution Across Diverse Tasks](https://mlsanity.com/paper/2502.14848/) (2025)
- [LLM Library Learning Fails: A LEGO-Prover Case Study](https://mlsanity.com/paper/2504.03048/) (2025)
- [Characterizing Language Use in a Collaborative Situated Game](https://mlsanity.com/paper/2512.03381/) (2025)
- [Transductive Visual Programming: Evolving Tool Libraries from Experience for Spatial Reasoning](https://mlsanity.com/paper/2512.20934/) (2025)
- [Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?](https://mlsanity.com/paper/2607.15439/) (2026)
- [AbstractBeam: Enhancing Bottom-Up Program Synthesis using Library Learning](https://mlsanity.com/paper/2405.17514/) (2024)
- [Self-Evolving Visual Concept Library using Vision-Language Critics](https://mlsanity.com/paper/2504.00185/) (2025)
- [A Compute-Matched Re-Evaluation of TroVE on MATH](https://mlsanity.com/paper/2507.22069/) (2025)

## Abstract (arXiv)

While large language models (LLMs) now excel at code generation, a key aspect of software development is the art of refactoring: consolidating code into libraries of reusable and readable programs. In this paper, we introduce LILO, a neurosymbolic framework that iteratively synthesizes, compresses, and documents code to build libraries tailored to particular problem domains. LILO combines LLM-guided program synthesis with recent algorithmic advances in automated refactoring from Stitch: a symbolic compression system that efficiently identifies optimal lambda abstractions across large code corpora. To make these abstractions interpretable, we introduce an auto-documentation (AutoDoc) procedure that infers natural language names and docstrings based on contextual examples of usage. In addition to improving human readability, we find that AutoDoc boosts performance by helping LILO's synthesizer to interpret and deploy learned abstractions. We evaluate LILO on three inductive program synthesis benchmarks for string editing, scene reasoning, and graphics composition. Compared to existing neural and symbolic methods - including the state-of-the-art library learning algorithm DreamCoder - LILO solves more complex tasks and learns richer libraries that are grounded in linguistic knowledge.