VisualCoder: Guiding Vision Language Models in Code Execution with Fine-grained Chain-of-Thought Reasoning

Cuong Le Chi,Chau Truong Vinh Hoang,Phan Nhật Huy,Dung D. Le,Tien N Nguyen,Nghi D. Q. Bui

提交: 2024-09-28更新: 2024-10-15

TL;DR

We present VISUALCODER, a method that enhances LLMs' reasoning on code execution by integrating Control Flow Graphs (CFGs) with Chain-of-Thought (CoT) reasoning.

摘要

关键词

Multimodal Large Language ModelsControl Flow GraphsVisualized Code RepresentationCode Execution ReasoningFault LocalizationProgram Repair

评审与讨论

撤稿通知

2024-10-15

I have read and agree with the venue's withdrawal policy on behalf of myself and my co-authors.