Authors: Xinyi He, Jiaru Zou, Yun Lin, Mengyu Zhou, Shi Han, Zejian Yuan, Dongmei Zhang

Published on: March 20, 2024

Impact Score: 7.6

Arxiv code: Arxiv:2403.13583

Summary

• What is new: Introduces CONLINE framework that incorporates online searches and automated correctness testing into code generation processes.
• Why this is important: Generating complex code with large language models remains challenging in real-world scenarios.
• What the research proposes: CONLINE framework, which enhances code generation through online information retrieval and iterative refinement with automated testing.
• Results: CONLINE significantly improves the quality of complex code generation, as shown through rigorous experiments on DS-1000 and ClassEval datasets.

Technical Details

Technological frameworks used: CONLINE

Models used: Large Language Models (LLMs)

Data used: DS-1000, ClassEval

Potential Impact

Software development companies, coding education platforms, and developer tool providers.

Want to implement this idea in a business?

We have generated a startup concept here: CodeFlux.