Rethinking Tabular Data Understanding with Large Language Models
This addresses the problem of improving LLM-based tabular data understanding for AI applications, representing a substantial but incremental advancement over existing methods.
The study investigated how large language models interpret tabular data, finding that structural variations in tables cause performance drops, especially in symbolic reasoning. By aggregating textual and symbolic reasoning pathways with a self-consistency mechanism, they achieved state-of-the-art performance with 73.6% accuracy on WIKITABLEQUESTIONS.
Large Language Models (LLMs) have shown to be capable of various tasks, yet their capability in interpreting and reasoning over tabular data remains an underexplored area. In this context, this study investigates from three core perspectives: the robustness of LLMs to structural perturbations in tables, the comparative analysis of textual and symbolic reasoning on tables, and the potential of boosting model performance through the aggregation of multiple reasoning pathways. We discover that structural variance of tables presenting the same content reveals a notable performance decline, particularly in symbolic reasoning tasks. This prompts the proposal of a method for table structure normalization. Moreover, textual reasoning slightly edges out symbolic reasoning, and a detailed error analysis reveals that each exhibits different strengths depending on the specific tasks. Notably, the aggregation of textual and symbolic reasoning pathways, bolstered by a mix self-consistency mechanism, resulted in achieving SOTA performance, with an accuracy of 73.6% on WIKITABLEQUESTIONS, representing a substantial advancement over previous existing table processing paradigms of LLMs.