Reproduce “Enhancing Complex Question Answering over Knowledge Graphs through Evidence Pattern Retrieval” 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

Wentao Ding, Jinmao Li, Liangchuan Luo, Yuzhong Qu

Nanjing University researchers developed Evidence Pattern Retrieval (EPR) to improve Knowledge Graph Question Answering by explicitly modeling structural dependencies among evidence facts during subgraph extraction. The method achieved a new state-of-the-art for Information Retrieval (IR) approaches on the ComplexWebQuestions benchmark, reaching 60.6% Hits@1 and 61.2% F1 score.

Key insights

  • Explicitly identifying and leveraging "evidence patterns," which represent structural dependencies, significantly enhances the quality of extracted subgraphs for complex questions.
  • Atomic pattern retrieval and iterative construction allow for efficient and comprehensive generation of candidate evidence patterns.
  • Providing a cleaner, more relevant subgraph through structural pattern matching drastically improves the performance of downstream answer reasoning modules.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.