DS@GT at CheckThat! 2025: Evaluating Context and Tokenization Strategies for Numerical Fact Verification

TL;DR

This paper evaluates different modeling strategies for numerical fact verification, focusing on context length and tokenization methods, and finds evidence quality to be the main bottleneck, achieving a top-4 system in the CheckThat! 2025 challenge.

Contribution

It systematically assesses the impact of context size and R2L tokenization on numerical claim verification, providing insights contrary to prior arithmetic reasoning studies.

Findings

Longer context windows do not improve performance.

R2L tokenization does not enhance numerical NLI.

Evidence quality is the primary bottleneck.

Abstract

Numerical claims, statements involving quantities, comparisons, and temporal references, pose unique challenges for automated fact-checking systems. In this study, we evaluate modeling strategies for veracity prediction of such claims using the QuanTemp dataset and building our own evidence retrieval pipeline. We investigate three key factors: (1) the impact of more evidences with longer input context windows using ModernBERT, (2) the effect of right-to-left (R2L) tokenization, and (3) their combined influence on classification performance. Contrary to prior findings in arithmetic reasoning tasks, R2L tokenization does not boost natural language inference (NLI) of numerical tasks. A longer context window does also not enhance veracity performance either, highlighting evidence quality as the dominant bottleneck. Our best-performing system achieves competitive macro-average F1 score of 0.57 and places us among the Top-4 submissions in Task 3 of CheckThat! 2025. Our code is available at https://github.com/dsgt-arc/checkthat-2025-numerical.