LGCLPLMar 25, 2022

CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis

CMU
arXiv:2203.13474v51485 citationsh-index: 112Has Code
Originality Highly original
AI Analysis

This work democratizes access to large language models for code generation by providing open-source models and a benchmark, addressing a resource limitation problem for researchers and developers in AI and programming.

The authors tackled program synthesis by training and releasing a family of large language models up to 16.1B parameters, called CODEGEN, which is competitive with previous state-of-the-art on zero-shot Python code generation on HumanEval, and they showed that using multi-turn prompts significantly improves synthesis over single-turn approaches.

Program synthesis strives to generate a computer program as a solution to a given problem specification, expressed with input-output examples or natural language descriptions. The prevalence of large language models advances the state-of-the-art for program synthesis, though limited training resources and data impede open access to such models. To democratize this, we train and release a family of large language models up to 16.1B parameters, called CODEGEN, on natural language and programming language data, and open source the training library JAXFORMER. We show the utility of the trained model by demonstrating that it is competitive with the previous state-of-the-art on zero-shot Python code generation on HumanEval. We further investigate the multi-step paradigm for program synthesis, where a single program is factorized into multiple prompts specifying subproblems. To this end, we construct an open benchmark, Multi-Turn Programming Benchmark (MTPB), consisting of 115 diverse problem sets that are factorized into multi-turn prompts. Our analysis on MTPB shows that the same intent provided to CODEGEN in multi-turn fashion significantly improves program synthesis over that provided as a single turn. We make the training library JAXFORMER and model checkpoints available as open source contribution: https://github.com/salesforce/CodeGen.

Code Implementations8 repos

Data from Papers with Code (CC-BY-SA-4.0)

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes