Effective Large Language Model Debugging with Best-first Tree Search

Jialin Song, Jonathan Raiman, Bryan Catanzaro

Large Language Models (LLMs) show promise in code generation tasks. However, their code-writing abilities are often limited in scope: while they can successfully implement simple f…