Search papers, labs, and topics across Lattice.
This paper introduces TabRank, a framework designed to enhance table re-ranking in structured information retrieval by leveraging chain-of-thought (CoT) reasoning. By training a compact reasoning model on a newly created dataset of 6728 reasoning traces, the authors demonstrate substantial performance improvements across multiple table retrieval datasets, achieving up to a 52.9% increase in accuracy on the TabFact subset. The results highlight TabRank's effectiveness in generalizing to complex multi-table reasoning scenarios, underscoring its potential to refine retrieval systems significantly.
TabRank boosts table retrieval accuracy by up to 52.9% through innovative chain-of-thought reasoning techniques.
The ability to retrieve relevant tables for answering questions is a key task for structured information retrieval. Multi-stage retrieval systems rely heavily on rerankers to refine candidate lists produced by efficient first-stage retrievers. As a result, neural rerankers and LLM-based reranking methods have become increasingly important due to their superior capacity for semantic understanding and reasoning compared to conventional sparse or dense retrieval models. Recently, Large Reasoning Models (LRMs) equipped with explicit chain-of-thought (CoT) reasoning have shown strong improvements in ranking quality in unstructured passage retrieval. In this work, we present TabRank, a framework for training reasoning rerankers for Tabular Retrieval. We first present a comprehensive dataset of 6728 reasoning traces for tabular reranking on the Natural Questions Tables dataset. We then explore two variants of training a compact reasoning model on these reasoning traces: explicit CoT distillation and conditioning the student reranker on the teacher's reasoning trace within the prompt. We stress-test TabRank on several out-of-distribution generalization settings on diverse domains and multi-table scenarios. Our approach significantly improves performance across a variety of table retrieval datasets, increasing Acc@10 by 30.5% on HybridQA, 15.2% on SQA, 52.9% on TabFact, and 13.1% on TATQA subsets of the Multi-Table QA Benchmark compared to the base model. Notably, TabRank generalizes effectively to multi-table reasoning. Our code, data and models are available at https://github.com/AdarshSingh7647/TabRanker
