IEEE

Fixing Broken Graphs: LLM-Powered Automatic Code Optimization for DNN Programs

Abstract: Deep learning compilers optimize DNN program execution by capturing them as operator-based computation graphs. However, developers’ deep learning programs often contain complex Python ...
IEEE

Line-Level Defect Prediction by Capturing Code Contexts With Graph Convolutional Networks

Abstract: Software defect prediction refers to the systematic analysis and review of software using various approaches and tools to identify potential defects or errors. Software defect prediction ...
GitHub

GraphGPT: Graph Instruction Tuning for Large Language Models

Data Intelligence Lab@University of Hong Kong, Baidu Inc. This repository hosts the code, data and model weight of GraphGPT (SIGIR'24 full paper track). Due to compatibility issues, if you are using ...
GitHub

CGBridge: Bridging Code Graphs and Large Language Models for Better Structure-Aware Code Understanding

CGBridge is a novel framework designed to enhance the code understanding capabilities of Large Language Models (LLMs) by integrating rich structural information from code graphs. Our approach follows ...