LLM-Based Test-Driven Interactive Code Generation: User Study and Empirical Evaluation

TL;DR

This paper introduces TiCoder, an interactive workflow that uses test-driven clarification to improve the accuracy of code generated by large language models, demonstrated through user studies and empirical tests.

Contribution

The paper presents a novel test-driven interactive workflow, TiCoder, that enhances LLM-based code generation accuracy and user understanding through intent clarification and automated testing.

Findings

Participants using TiCoder more accurately evaluated generated code.

TiCoder significantly reduced task-induced cognitive load.

Achieved an average 45.97% improvement in code accuracy across models and datasets.

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, given NL is informal, it does not lend easily to checking that the generated code correctly satisfies the user intent. In this paper, we propose a novel interactive workflow TiCoder for guided intent clarification (i.e., partial formalization) through tests to support the generation of more accurate code suggestions. Through a mixed methods user study with 15 programmers, we present an empirical evaluation of the effectiveness of the workflow to improve code generation accuracy. We find that participants using the proposed workflow are significantly more likely to correctly evaluate AI generated code, and report significantly less task-induced cognitive load. Furthermore, we test the potential of the workflow at scale with four different state-of-the-art LLMs on two python datasets, using an idealized proxy for a user feedback. We observe an average absolute improvement of 45.97% in the pass@1 code generation accuracy for both datasets and across all LLMs within 5 user interactions, in addition to the automatic generation of accompanying unit tests.