IMRNNs: An Efficient Method for Interpretable Dense Retrieval via Embedding Modulation

TL;DR

IMRNNs introduce a lightweight, dynamic modulation framework for dense retrieval that enhances interpretability and improves effectiveness by adaptively refining query and document embeddings during inference.

Contribution

The paper presents IMRNNs, a novel modular approach that dynamically modulates embeddings for better interpretability and retrieval performance in dense retrievers.

Findings

IMRNNs improve retrieval metrics by over 6% on benchmark datasets.

Dynamic embedding modulation enhances interpretability of query-document interactions.

The method achieves state-of-the-art results with minimal additional computational cost.

Abstract

Interpretability in black-box dense retrievers remains a central challenge in Retrieval-Augmented Generation (RAG). Understanding how queries and documents semantically interact is critical for diagnosing retrieval behavior and improving model design. However, existing dense retrievers rely on static embeddings for both queries and documents, which obscures this bidirectional relationship. Post-hoc approaches such as re-rankers are computationally expensive, add inference latency, and still fail to reveal the underlying semantic alignment. To address these limitations, we propose Interpretable Modular Retrieval Neural Networks (IMRNNs), a lightweight framework that augments any dense retriever with dynamic, bidirectional modulation at inference time. IMRNNs employ two independent adapters: one conditions document embeddings on the current query, while the other refines the query embedding using corpus-level feedback from initially retrieved documents. This iterative modulation process enables the model to adapt representations dynamically and expose interpretable semantic dependencies between queries and documents. Empirically, IMRNNs not only enhance interpretability but also improve retrieval effectiveness. Across seven benchmark datasets, applying our method to standard dense retrievers yields average gains of +6.35% nDCG, +7.14% recall, and +7.04% MRR over state-of-the-art baselines. These results demonstrate that incorporating interpretability-driven modulation can both explain and enhance retrieval in RAG systems.