PRISM: Agentic Retrieval with LLMs for Multi-Hop Question Answering

ICLR 2026 Conference Submission20470 Authors

Keywords: Information Retrieval, Question Answering, Retrieval-Augmented Generation

TL;DR: We propose an agentic retrieval framework for multi-hop QA that balances precision and recall via iterative LLM agents, yielding compact evidence, outperforming baselines, and boosting QA accuracy with less irrelevant context.

Abstract: 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.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 20470