RASL retrieves relevant tables and columns for text-to-SQL by decomposing schemas into semantic entities, calibrating entity-type importance on training data, and using an LLM to rank candidates, beating baselines on Spider, BIRD, and Fiben.
Is Table Retrieval a Solved Problem? Exploring Join-Aware Multi-Table Retrieval
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Retrieving relevant tables containing the necessary information to accurately answer a given question over tables is critical to open-domain question-answering (QA) systems. Previous methods assume the answer to such a question can be found either in a single table or multiple tables identified through question decomposition or rewriting. However, neither of these approaches is sufficient, as many questions require retrieving multiple tables and joining them through a join plan that cannot be discerned from the user query itself. If the join plan is not considered in the retrieval stage, the subsequent steps of reasoning and answering based on those retrieved tables are likely to be incorrect. To address this problem, we introduce a method that uncovers useful join relations for any query and database during table retrieval. We use a novel re-ranking method formulated as a mixed-integer program that considers not only table-query relevance but also table-table relevance that requires inferring join relationships. Our method outperforms the state-of-the-art approaches for table retrieval by up to 9.3% in F1 score and for end-to-end QA by up to 5.4% in accuracy.
citation-role summary
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
RASL: Retrieval Augmented Schema Linking for Massive Database Text-to-SQL
RASL retrieves relevant tables and columns for text-to-SQL by decomposing schemas into semantic entities, calibrating entity-type importance on training data, and using an LLM to rank candidates, beating baselines on Spider, BIRD, and Fiben.