Foresight method improves question answering over knowledge graphs
Foresight-over-Graph: Reasoning Beyond Local Horizons for Knowledge Base Question Answering
Computation and LanguageArtificial Intelligence
Summary
Large language models sometimes make up facts when answering questions that need a lot of real-world knowledge. The authors connect these models to knowledge graphs, which store real facts in a structured way, to help them reason better. Usually, finding right facts step-by-step can miss important connections if those facts don’t look useful at first. The authors propose a new way called Foresight-over-Graph that looks further ahead when choosing facts, leading to better answers with fewer calls to the language model. They tested this on standard benchmarks and saw large improvements in accuracy and efficiency.
What this means in practice
- •For software developers building question answering systems: Improve accuracy and reduce costs in question answering over knowledge graphs by using a foresight-based retrieval method to find relevant evidence paths.
- •For enterprise knowledge management teams: Enhance reliability of automated answers by integrating foresight-guided graph reasoning that avoids missing critical evidence branches in complex knowledge bases.
Authors
Yang Hong, Yajun Yang, Xin Wang, Liping Jing, Qinghua Hu
Abstract
Large language models (LLMs) have demonstrated strong capabilities in question answering, yet they still frequently suffer from hallucinations on knowledge-intensive tasks. Knowledge graphs (KGs) provide LLMs with structured, interpretable, and updatable factual grounding, making them a promising external knowledge source for reliable reasoning. However, existing LLM-guided graph reasoning methods typically rely on hop-wise greedy or beam-style pruning during evidence retrieval. Such local decision processes are inherently myopic: evidence that appears weak near the source may become crucial only after deeper graph context is explored, causing answer-critical branches to be discarded prematurely and making the reasoning chain difficult to recover. To address this limitation, we propose Foresight-over-Graph (FoG), a foresight-aware evidence retrieval framework for knowledge base question answering (KBQA). FoG iteratively constructs a question-relevant evidence subgraph and uses far-to-near feedback to guide path exploration, and maintains a compact memory subgraph to support continued exploration. Extensive experiments on widely used KBQA benchmarks demonstrate that FoG achieves state-of-the-art performance, with a particularly large improvement of 16.58% in Hit on CWQ, while also reducing LLM calls and token usage. Our code is available at https://github.com/yhong7/FoG .