Retrieval--Reasoning Processes for Multi-hop Question Answering: A Four-Axis Design Framework and Empirical Trends

TL;DR

This paper introduces a four-axis framework to analyze retrieval and reasoning processes in multi-hop question answering systems, highlighting procedural choices and trends across models and benchmarks.

Contribution

It provides a novel four-axis schema for analyzing multi-hop QA systems' execution processes and synthesizes empirical trends and challenges in the field.

Findings

Mapping of representative systems using the framework

Identification of trade-offs among effectiveness, efficiency, and faithfulness

Highlighting open challenges like structure-aware planning and robust stopping

Abstract

Multi-hop question answering (QA) requires systems to iteratively retrieve evidence and reason across multiple hops. While recent RAG and agentic methods report strong results, the underlying retrieval--reasoning \emph{process} is often left implicit, making procedural choices hard to compare across model families. This survey takes the execution procedure as the unit of analysis and introduces a four-axis framework covering (A) overall execution plan, (B) index structure, (C) next-step control (strategies and triggers), and (D) stop/continue criteria. Using this schema, we map representative multi-hop QA systems and synthesize reported ablations and tendencies on standard benchmarks (e.g., HotpotQA, 2WikiMultiHopQA, MuSiQue), highlighting recurring trade-offs among effectiveness, efficiency, and evidence faithfulness. We conclude with open challenges for retrieval--reasoning agents, including structure-aware planning, transferable control policies, and robust stopping under distribution shift.