HyperProve improves multi-step question answering with answer-guided search
HyperProve: Answer-Guided Hypergraph Expansion for Multi-Hop Question Answering
Computation and Language
Summary
Answering complex questions often requires piecing together multiple facts that are connected. The authors introduce HyperProve, a method that breaks down questions and builds chains of related facts step-by-step, keeping track of possible answers along the way. This helps gather more relevant evidence and improves the accuracy of final answers. Their approach performed better than strong existing methods on several benchmark tests.
What this means in practice
- •For natural language processing teams: Build question answering systems that retrieve and combine multi-step evidence more accurately to improve complex query responses.
- •For digital assistant developers: Enhance virtual assistants to handle complex user queries by constructing and validating multi-fact answers more reliably.
Authors
An Nguyen Phu, Dung Nguyen Quang, Luu Hieu An, Linh Ngo Van, Trung Le, Thien Huu Nguyen
Abstract
Multi-hop question answering often fails when retrieval treats evidence as isolated matches to the original question, since the facts needed to answer a complex question are usually connected through intermediate entities, relations, and constraints. We propose HyperProve, a retrieval-augmented QA framework that addresses this challenge by coupling question decomposition with answer-conditioned expansion over a hypergraph of atomic facts. HyperProve does not use atomic facts, hypergraphs, or iterative retrieval in isolation; instead, it carries intermediate answers and supporting hyperedges as retrieval state, then uses that state to bias the next local hypergraph expansion. This design enables HyperProve to construct coherent evidence chains for final answer generation while making the retrieval process stateful and fact-centered. Across multi-hop QA benchmarks, HyperProve achieves the best overall performance in our evaluation, outperforming the strongest baselines by an average relative improvement of 6.2% in answer accuracy and 4.9% in F1.