Interpreting Conversational Dense Retrieval By Rewriting-enhanced Inversion Of Session Embedding

Abstract

Conversational dense retrieval has shown to be effective in conversational search. However, a major limitation of conversational dense retrieval is their lack of interpretability, hindering intuitive understanding of model behaviors for targeted improvements. This paper presents CONVINV, a simple yet effective approach to shed light on interpretable conversational dense retrieval models. CONVINV transforms opaque conversational session embeddings into explicitly interpretable text while faithfully maintaining their original retrieval performance as much as possible. Such transformation is achieved by training a recently proposed Vec2Text model based on the ad-hoc query encoder, leveraging the fact that the session and query embeddings share the same space in existing conversational dense retrieval. To further enhance interpretability, we propose to incorporate external interpretable query rewrites into the transformation process. Extensive evaluations on three conversational search ben

Stats

• citations7
• S2 citations—
• github stars0
• HF likes0
• heat score6.77
• arxiv keycheng2024interpreting

Related papers

• Contextualized Query Embeddings For Conversational Search (2021)10.21
• CONVERSER: Few-shot Conversational Dense Retrieval With Synthetic Data Generation (2023)8.25
• Few-shot Conversational Dense Retrieval (2021)16.68
• Dense Passage Retrieval In Conversational Search (2025)0.00
• Imrnns: An Efficient Method For Interpretable Dense Retrieval Via Embedding Modulation (2026)0.00
• Interpret And Control Dense Retrieval With Sparse Latent Features (2024)2.26
• What Are You Token About? Dense Retrieval As Distributions Over The Vocabulary (2022)8.09
• Understanding And Mitigating The Threat Of Vec2text To Dense Retrieval Systems (2024)7.16