Evidence-Guided Schema Normalization for Temporal Tabular Reasoning

TL;DR

This paper introduces an evidence-guided schema normalization approach for temporal reasoning in semi-structured tables, demonstrating that schema quality significantly improves question-answering accuracy more than model size.

Contribution

It proposes a novel schema normalization method based on evidence principles, significantly enhancing QA performance over existing models.

Findings

Schema normalization greatly improves QA accuracy.

Semantic naming reduces ambiguity in schemas.

Best configuration achieves 80.39 EM, 16.8% above baseline.

Abstract

Temporal reasoning over evolving semi-structured tables poses a challenge to current QA systems. We propose a SQL-based approach that involves (1) generating a 3NF schema from Wikipedia infoboxes, (2) generating SQL queries, and (3) query execution. Our central finding challenges model scaling assumptions: the quality of schema design has a greater impact on QA precision than model capacity. We establish three evidence-based principles: normalization that preserves context, semantic naming that reduces ambiguity, and consistent temporal anchoring. Our best configuration (Gemini 2.5 Flash schema + Gemini-2.0-Flash queries) achieves 80.39 EM, a 16.8\% improvement over the baseline (68.89 EM).