UniRetriever: Multi-task Candidates Selection for Various Context-Adaptive Conversational Retrieval

TL;DR

UniRetriever is a multi-task, universal retrieval model designed to efficiently select persona, knowledge, and response candidates in conversational systems, outperforming previous specialized retrievers and demonstrating strong generalization.

Contribution

The paper introduces a dual-encoder multi-task framework that unifies retrieval tasks in conversational AI, improving efficiency and performance over independent retrievers.

Findings

Achieves state-of-the-art retrieval quality within and outside training domains.

Effectively models relationships between dialogue context and candidates.

Demonstrates strong generalization as a universal retriever.

Abstract

Conversational retrieval refers to an information retrieval system that operates in an iterative and interactive manner, requiring the retrieval of various external resources, such as persona, knowledge, and even response, to effectively engage with the user and successfully complete the dialogue. However, most previous work trained independent retrievers for each specific resource, resulting in sub-optimal performance and low efficiency. Thus, we propose a multi-task framework function as a universal retriever for three dominant retrieval tasks during the conversation: persona selection, knowledge selection, and response selection. To this end, we design a dual-encoder architecture consisting of a context-adaptive dialogue encoder and a candidate encoder, aiming to attention to the relevant context from the long dialogue and retrieve suitable candidates by simply a dot product. Furthermore, we introduce two loss constraints to capture the subtle relationship between dialogue context and different candidates by regarding historically selected candidates as hard negatives. Extensive experiments and analysis establish state-of-the-art retrieval quality both within and outside its training domain, revealing the promising potential and generalization capability of our model to serve as a universal retriever for different candidate selection tasks simultaneously.