LLM-Symbolic Integration for Robust Temporal Tabular Reasoning

TL;DR

This paper introduces TempTabQA-C, a synthetic dataset and a symbolic approach enabling LLMs to perform robust, scalable temporal reasoning over tables by generating and executing SQL queries, outperforming previous methods.

Contribution

The paper presents a novel symbolic intermediate representation and adaptive few-shot prompting to improve LLMs' temporal reasoning over structured data.

Findings

Enhanced robustness and scalability in temporal reasoning tasks.

Significant performance improvements over traditional prompting methods.

Set a new benchmark for LLM-based temporal tabular reasoning.

Abstract

Temporal tabular question answering presents a significant challenge for Large Language Models (LLMs), requiring robust reasoning over structured data, which is a task where traditional prompting methods often fall short. These methods face challenges such as memorization, sensitivity to table size, and reduced performance on complex queries. To overcome these limitations, we introduce TempTabQA-C, a synthetic dataset designed for systematic and controlled evaluations, alongside a symbolic intermediate representation that transforms tables into database schemas. This structured approach allows LLMs to generate and execute SQL queries, enhancing generalization and mitigating biases. By incorporating adaptive few-shot prompting with contextually tailored examples, our method achieves superior robustness, scalability, and performance. Experimental results consistently highlight improvements across key challenges, setting a new benchmark for robust temporal reasoning with LLMs.