DOTRAG: Retrieval-Time Reasoning Along Paths

TL;DR

DotRAG introduces a retrieval framework that guides graph exploration through reasoning-based constraints, improving multi-hop reasoning performance without additional training.

Contribution

It reformulates retrieval as a reasoning process over paths, enabling adaptive, query-specific exploration without explicit step-by-step reasoning chains.

Findings

Achieves state-of-the-art on MetaQA and UltraDomain datasets.

Demonstrates consistent improvements on multi-hop tasks.

Operates without additional training or explicit reasoning chains.

Abstract

Graph Retrieval-Augmented Generation (GraphRAG) is dominated by a retrieve-then-reason paradigm, where context is retrieved using heuristics and then reasoned over. Such methods struggle to adapt to the query-specific logic required for complex multi-hop tasks, often accumulating irrelevant context or missing correct relational paths. We propose DotRAG, a training-free GraphRAG framework that reformulates retrieval as a reasoning process over paths. Our approach generates query-conditioned constraints that guide graph exploration, prune irrelevant regions, and iteratively discover relational paths without relying on explicit step-by-step reasoning chains. We introduce Division of Thought (DOT), an abstraction that decomposes retrieval into localized search spaces and adapts the search strategy to each query. DotRAG achieves SOTA performance on MetaQA and UltraDomain, with consistent gains on multi-hop tasks, demonstrating the effectiveness of reasoning-guided retrieval.