Reproduce “Evolve on Graph: Self-Evolving Knowledge Graph Reasoning without Question Annotations” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zongyuan Sun, Shuai Zhao, Ke Shu, Ziyue Zhu, Shangyang Wu, Yuanlin Lin, Tao Wang, Shiyue Li and 3 more
EoG trains a language model to explore knowledge graphs without annotated question–answer training labels. A Proposer generates graph-grounded questions, executed graph queries supply answer sets, and a Solver’s performance guides difficulty. On five public benchmarks, Qwen-2.5-7B-Instruct gains 7.46 Hit@1 and 12.05 F1 points over Graph Reranker on average; on IndustryCarBench, it beats ToG by 17.80 Hit@1 points.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.