Ask Optimal Questions: Aligning Large Language Models with Retriever’s Preference in Conversational Search
Abstract: Conversational search, unlike single-turn retrieval tasks, requires understanding the current question within a dialogue context.
The common approach of *rewrite-then-retrieve* aims to decontextualize questions to be self-sufficient for off-the-shelf retrievers, but most existing methods produce sub-optimal query rewrites due to the limited ability to incorporate signals from the retrieval results.
To overcome this limitation, we present a novel framework ***RetPO*** (**Ret**riever's **P**reference **O**ptimization), which is designed to optimize a language model (LM) for reformulating search queries in line with the preferences of the target retrieval systems.
The process begins by prompting a large LM to produce various potential rewrites and then collects retrieval performance for these rewrites as the retrievers' preferences.
Through the process, we construct a large-scale dataset called ***RF collection***, containing **R**etrievers' **F**eedback on over 410K query rewrites across 12K conversations.
Furthermore, we fine-tune a smaller LM using this dataset to align it with the retrievers' preferences as feedback.
The resulting model demonstrates superiority on two benchmarks, surpassing the previous state-of-the-art performance of *rewrite-then-retrieve* approaches, including GPT-3.5. Code and dataset will be available.
Paper Type: Long
Research Area: Information Retrieval and Text Mining
Research Area Keywords: passage retrieval, dense retrieval, query reformulation
Contribution Types: NLP engineering experiment, Approaches to low-resource settings, Publicly available software and/or pre-trained models, Data resources
Languages Studied: English
Submission Number: 4029
Loading