ConvGQR: Generative Query Reformulation for Conversational Search

• Paper: https://arxiv.org/abs/2305.15645
• Authors: Fengran Mo, Kelong Mao, Yutao Zhu, Yihong Wu, Kaiyu Huang, Jian-Yun Nie
• Published: 2023-05-25
• Topics: Conversational Search, Multi-Turn Retrieval, Query Understanding

Problem

In conversational search, the user's real search intent for the current turn depends on the previous conversation history. Determining a good search query from the whole conversation context is challenging. To avoid expensive re-training of the query encoder, most existing methods learn a rewriting model that de-contextualizes the current query by mimicking manual query rewriting. However, manually rewritten queries are not always the best search queries, and training a rewriting model on them limits its ability to produce good search queries. Another useful hint is the potential answer to the question.

Approach

ConvGQR reformulates conversational queries using generative pre-trained language models (PLMs) with two components:

1. Query rewriting module — de-contextualizes the current query given the conversation history. 2. Answer generation module — generates a potential answer to the current question.

By combining the rewritten query with the generated potential answer, ConvGQR produces better search queries than rewriting alone.

Additionally, a knowledge infusion mechanism relates query reformulation to retrieval performance, jointly optimizing both the reformulation and the retrieval.

Results

Extensive experiments on four conversational search datasets demonstrate the effectiveness of ConvGQR over existing query rewriting baselines.

Original Abstract

> In conversational search, the user's real search intent for the current turn is dependent on the previous conversation history. It is challenging to determine a good search query from the whole conversation context. To avoid the expensive re-training of the query encoder, most existing methods try to learn a rewriting model to de-contextualize the current query by mimicking the manual query rewriting. However, manually rewritten queries are not always the best search queries. Training a rewriting model on them would limit the model's ability to produce good search queries. Another useful hint is the potential answer to the question. In this paper, we propose ConvGQR, a new framework to reformulate conversational queries based on generative pre-trained language models (PLMs), one for query rewriting and another for generating potential answers. By combining both, ConvGQR can produce better search queries. In addition, to relate query reformulation to retrieval performance, we propose a knowledge infusion mechanism to optimize both query reformulation and retrieval. Extensive experiments on four conversational search datasets demonstrate the effectiveness of ConvGQR.

Related Entries

• A Survey of Conversational Search (ACM, Sep 2025)
• Engineering Conversational Search Systems: A Review (arXiv 2407.00997)
• Agentic Conversational Search with Contextualized Reasoning via Reinforcement Learning
• CTR-Guided Generative Query Suggestion in Conversational Search (EMNLP)
• ChatRetriever: Adapting LLMs for Generalized and Robust Conversational Search (arXiv 2404.13556)
• CoSearchAgent: A Lightweight Collaborative Search Agent (arXiv 2402.06360)

Glossary

| Term | Meaning | |------|---------| | IR | Information Retrieval | | RAG | Retrieval-Augmented Generation | | PLM | Pre-trained Language Model | | nDCG | Normalized Discounted Cumulative Gain, a ranking quality metric | | Query Rewriting | De-contextualizing a conversational query into a standalone search query |