PRISM: Agentic Retrieval with LLMs for Multi-Hop Question Answering
This addresses the problem of retrieving relevant evidence for complex multi-hop questions, improving accuracy for QA systems, though it is incremental as it builds on existing retrieval and LLM methods.
The paper tackles multi-hop question answering by introducing an agentic retrieval system that uses LLMs in a structured loop to decompose questions and retrieve evidence with high precision and recall, achieving higher retrieval accuracy and enabling downstream QA models to surpass full-context answer accuracy on four benchmarks.
Retrieval plays a central role in multi-hop question answering (QA), where answering complex questions requires gathering multiple pieces of evidence. We introduce an Agentic Retrieval System that leverages large language models (LLMs) in a structured loop to retrieve relevant evidence with high precision and recall. Our framework consists of three specialized agents: a Question Analyzer that decomposes a multi-hop question into sub-questions, a Selector that identifies the most relevant context for each sub-question (focusing on precision), and an Adder that brings in any missing evidence (focusing on recall). The iterative interaction between Selector and Adder yields a compact yet comprehensive set of supporting passages. In particular, it achieves higher retrieval accuracy while filtering out distracting content, enabling downstream QA models to surpass full-context answer accuracy while relying on significantly less irrelevant information. Experiments on four multi-hop QA benchmarks -- HotpotQA, 2WikiMultiHopQA, MuSiQue, and MultiHopRAG -- demonstrates that our approach consistently outperforms strong baselines.