Multi hop question answering improved by smart evidence retrieval method
Beyond One-Shot Expansion: Contrastive Evidence Exploration for Multi-Hop Retrieval
Artificial IntelligenceInformation Retrieval
Summary
Finding the right bits of information to answer complex questions can be tricky when the answer requires combining multiple facts. The authors created a new way to find helpful passages step-by-step by using clues from what has already been found, instead of trying to find everything at once. Their method also compares passages carefully to pick out the most useful ones without repeating similar information. Tests showed this approach helps computers find better evidence and answer questions more accurately.
retrieval augmented generationmulti hop question answeringevidence retrievalquery expansioncontrastive refinementoffline indexingpassage rankingMuSiQueHotpotQA2WikiMultihopQA
Authors
JungMin Yun, YoungBin Kim
Abstract
Retrieval-augmented generation (RAG) critically depends on retrieving the evidence necessary for effective reasoning. However, this remains particularly challenging in multi-hop question answering (QA), where supporting passages are often linked through intermediate entities and relations that must be progressively uncovered. Existing retrieval approaches typically rely on a single retrieval intent or one-shot query expansion, limiting their ability to adapt to newly retrieved evidence and potentially introducing noisy or redundant retrieval signals. To address these limitations, we propose a training-free multi-hop retrieval framework that integrates evidence-conditioned exploration, passage-specific contrastive refinement, and coverage-aware final ranking. During offline indexing, the framework constructs passage-specific contrastive facets that characterize each passage relative to its semantically similar neighbors, providing fine-grained signals to distinguish closely related candidates. At inference time, the framework iteratively retrieves evidence, generates probes targeting unresolved information needs, refines candidate relevance using the contrastive facets, and selects a complementary set of passages that collectively cover diverse evidence-seeking intents. Experiments on MuSiQue, HotpotQA, and 2WikiMultihopQA demonstrate consistent improvements in retrieval quality and downstream QA performance over baselines.