LILO: Learning Interpretable Libraries by Compressing and Documenting Code
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916159966674944 |
|---|---|
| author | Grand, Gabriel Wong, Lionel Bowers, Maddy Olausson, Theo X. Liu, Muxin Tenenbaum, Joshua B. Andreas, Jacob |
| author_facet | Grand, Gabriel Wong, Lionel Bowers, Maddy Olausson, Theo X. Liu, Muxin Tenenbaum, Joshua B. Andreas, Jacob |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_19791 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | LILO: Learning Interpretable Libraries by Compressing and Documenting Code Grand, Gabriel Wong, Lionel Bowers, Maddy Olausson, Theo X. Liu, Muxin Tenenbaum, Joshua B. Andreas, Jacob Computation and Language Artificial Intelligence Machine Learning Programming Languages 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. |
| title | LILO: Learning Interpretable Libraries by Compressing and Documenting Code |
| topic | Computation and Language Artificial Intelligence Machine Learning Programming Languages |
| url | https://arxiv.org/abs/2310.19791 |