ComposeRAG: A Modular and Composable RAG for Corpus-Grounded Multi-Hop Question Answering

TL;DR

ComposeRAG introduces a modular, composable framework for multi-hop question answering that improves accuracy, grounding, and interpretability by decomposing RAG pipelines into atomic modules with self-reflection capabilities.

Contribution

It presents a novel modular abstraction for RAG systems, enabling independent module development, targeted improvements, and enhanced robustness in multi-hop QA tasks.

Findings

Up to 15% accuracy improvement over fine-tuning methods

Reduces ungrounded answers by over 10% in low-quality retrieval settings

Significantly improves grounding fidelity across benchmarks

Abstract

Retrieval-Augmented Generation (RAG) systems are increasingly diverse, yet many suffer from monolithic designs that tightly couple core functions like query reformulation, retrieval, reasoning, and verification. This limits their interpretability, systematic evaluation, and targeted improvement, especially for complex multi-hop question answering. We introduce ComposeRAG, a novel modular abstraction that decomposes RAG pipelines into atomic, composable modules. Each module, such as Question Decomposition, Query Rewriting, Retrieval Decision, and Answer Verification, acts as a parameterized transformation on structured inputs/outputs, allowing independent implementation, upgrade, and analysis. To enhance robustness against errors in multi-step reasoning, ComposeRAG incorporates a self-reflection mechanism that iteratively revisits and refines earlier steps upon verification failure. Evaluated on four challenging multi-hop QA benchmarks, ComposeRAG consistently outperforms strong baselines in both accuracy and grounding fidelity. Specifically, it achieves up to a 15% accuracy improvement over fine-tuning-based methods and up to a 5% gain over reasoning-specialized pipelines under identical retrieval conditions. Crucially, ComposeRAG significantly enhances grounding: its verification-first design reduces ungrounded answers by over 10% in low-quality retrieval settings, and by approximately 3% even with strong corpora. Comprehensive ablation studies validate the modular architecture, demonstrating distinct and additive contributions from each component. These findings underscore ComposeRAG's capacity to deliver flexible, transparent, scalable, and high-performing multi-hop reasoning with improved grounding and interpretability.