Enhancing Complex Question Answering over Knowledge Graphs through Evidence Pattern Retrieval
Preprint
Abstract
Information retrieval (IR) methods for KGQA consist of two stages: subgraph extraction and answer reasoning. We argue current subgraph extraction methods underestimate the importance of structural dependencies among evidence facts. We propose Evidence Pattern Retrieval (EPR) to explicitly model the structural dependencies during subgraph extraction. We implement EPR by indexing the atomic adjacency pattern of resource pairs. Given a question, we perform dense retrieval to obtain atomic patterns formed by resource pairs. We then enumerate their combinations to construct candidate evidence patterns. These evidence patterns are scored using a neural model, and the best one is selected to extract a subgraph for downstream answer reasoning. Experimental results demonstrate that the EPR-based approach has significantly improved the F1 scores of IR-KGQA methods by over 10 points on ComplexWebQuestions and achieves competitive performance on WebQuestionsSP.Keywords
All Related Versions
This publication has 23 references indexed in Scilit:
- Code-Style In-Context Learning for Knowledge-Based Question AnsweringProceedings of the AAAI Conference on Artificial Intelligence, 2024
- FlexKBQA: A Flexible LLM-Powered Framework for Few-Shot Knowledge Base Question AnsweringProceedings of the AAAI Conference on Artificial Intelligence, 2024
- Complex Knowledge Base Question Answering: A SurveyIEEE Transactions on Knowledge and Data Engineering, 2022
- Beyond I.I.D.: Three Levels of Generalization for Question Answering on Knowledge Bases2021
- Improving Multi-hop Knowledge Base Question Answering by Learning Intermediate Supervision Signals2021
- Dense Passage Retrieval for Open-Domain Question Answering2020
- LC-QuAD 2.0: A Large Dataset for Complex Question Answering over Wikidata and DBpedia2019
- Billion-Scale Similarity Search with GPUsIEEE Transactions on Big Data, 2019
- An Introduction to Question Answering over Linked Data2014
- Freebase2008