Interactive Code Generation via Test-Driven User-Intent Formalization

Abstract

Large language models (LLMs) have shown great potential in automating significant aspects of coding by producing natural code from informal natural language (NL) intent. However, when interacting with LLMs, users have no guarantees that the code suggestions produced correctly satisfy the intent they provided. In fact, it is hard to define a notion of correctness since natural language can be ambiguous and lacks a formal semantics. In this paper, we propose the workflow of {\it interactive test-driven code generation}, which leverages lightweight user feedback to (a) formalize the user intent using generated tests that can be useful for debugging, and (b) produce an improved set of code suggestions by pruning and ranking candidate code suggestions. We describe a language-agnostic abstract algorithm and a concrete implementation TiCoder. We perform an automated evaluation of TiCoder on the \emph{MBPP} and \emph{HumanEval} code generation benchmarks. Our results are promising with using the OpenAI Codex LLM: our best algorithm improves the \passk{1} code generation accuracy (in absolute percentages) between $22.49\%$ to $37.71\%$ for MBPP and between $24.79\%$ to $53.98\%$ for HumanEval using between 1 to 5 simulated user queries.

18 pages

Cited by in corpus (6)

• CoverUp: Effective High Coverage Test Generation for Python
• Incremental Learning of Humanoid Robot Behavior from Natural Interaction and Large Language Models
• What Should We Engineer in Prompts? Training Humans in Requirement-Driven LLM Use
• Validating AI-Generated Code with Live Programming
• Generative AI for Test Driven Development: Preliminary Results
• Mokav: Execution-driven Differential Testing with LLMs