SUQL: Conversational Search over Structured and Unstructured Data with Large Language Models
It addresses the challenge of building conversational agents for real-world knowledge bases that mix structured and unstructured data, which is an incremental advancement over existing methods focused on single data types.
The paper tackles the problem of conversational search over hybrid structured and unstructured data by introducing SUQL, a language that extends SQL with free-text primitives, and an LLM-based semantic parser. The approach achieves within 8.9% exact match of SOTA on HybridQA and 90.3% success rate on a new Yelp dataset, compared to 63.4% for a baseline.
While most conversational agents are grounded on either free-text or structured knowledge, many knowledge corpora consist of hybrid sources. This paper presents the first conversational agent that supports the full generality of hybrid data access for large knowledge corpora, through a language we developed called SUQL (Structured and Unstructured Query Language). Specifically, SUQL extends SQL with free-text primitives (summary and answer), so information retrieval can be composed with structured data accesses arbitrarily in a formal, succinct, precise, and interpretable notation. With SUQL, we propose the first semantic parser, an LLM with in-context learning, that can handle hybrid data sources. Our in-context learning-based approach, when applied to the HybridQA dataset, comes within 8.9% exact match and 7.1% F1 of the SOTA, which was trained on 62K data samples. More significantly, unlike previous approaches, our technique is applicable to large databases and free-text corpora. We introduce a dataset consisting of crowdsourced questions and conversations on Yelp, a large, real restaurant knowledge base with structured and unstructured data. We show that our few-shot conversational agent based on SUQL finds an entity satisfying all user requirements 90.3% of the time, compared to 63.4% for a baseline based on linearization.