GitHub

GraphTranslator: Aligning Graph Model to Large Language Model for Open-ended Tasks

Pre-training Graph Model Phase. In the pre-training phase, we employ link prediction as the self-supervised task for pre-training the graph model. Producer Phase. In the Producer phase, we employ LLM ...
GitHub

Source code of our NeurIPS 2025 paper "Unifying and Enhancing Graph Transformers via a Hierarchical Mask Framework"

python==3.10.14 torch==2.3.1+cu121 --index-url https://download.pytorch.org/whl/cu121 ogb==1.3.6 torch-geometric==2.5.3 torch-scatter==2.1.2+pt23cu121 --index-url ...
IEEE

A Multi-Structural Graph Fusion Approach for Code Representation in Code Search

Abstract: Code search aims to retrieve semantically relevant code snippets on the basis of natural language queries. With the rapid expansion of public code repositories such as GitHub and Gitee, the ...
The Sun

Code breakers can win GOLD BAR worth £29,000 by solving tough puzzles – can YOU crack the Great British Treasure Hunt?

CODE breakers could win a gold bar worth up to £29,000 by cracking a series of ground-breaking challenges. The Royal Mint has launched a brand new £5 coin that’s used to solve a series of tough ...
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 ...