Beyond Independent Passages: Adaptive Passage Combination Retrieval for Retrieval Augmented Open-Domain Question Answering

TL;DR

This paper introduces AdaPCR, a novel retrieval framework that models dependencies between passages to improve open-domain question answering, especially for multi-hop questions, by adaptively selecting passage combinations for better context and answer accuracy.

Contribution

AdaPCR explicitly models inter-passage dependencies and adaptively retrieves passage combinations, advancing retrieval-augmented generation for open-domain QA.

Findings

Outperforms baseline methods on multiple QA benchmarks.

Significantly improves multi-hop reasoning accuracy.

Effectively models passage dependencies for better retrieval.

Abstract

Retrieval-augmented generation (RAG) enhances large language models (LLMs) by incorporating external documents at inference time, enabling up-to-date knowledge access without costly retraining. However, conventional RAG methods retrieve passages independently, often leading to redundant, noisy, or insufficiently diverse context-particularly problematic - particularly problematic in noisy corpora and for multi-hop questions. To address this, we propose Adaptive Passage Combination Retrieval (AdaPCR), a novel framework for open-domain question answering with black-box LMs. AdaPCR explicitly models dependencies between passages by considering passage combinations as units for retrieval and reranking. It consists of a context-aware query reformulation using concatenated passages, and a reranking step trained with a predictive objective aligned with downstream answer likelihood. Crucially, AdaPCR adaptively selects the number of retrieved passages without additional stopping modules. Experiments across several QA benchmarks show that AdaPCR outperforms baselines, particularly in multi-hop reasoning, demonstrating the effectiveness of modeling inter-passage dependencies for improved retrieval.