Abstract
Large language models (LLMs) are widely used for claim verification, yet remain brittle for numerical reasoning: even small changes in value can sharply degrade accuracy. We show that this brittleness persists in frontier LLMs, but can be mitigated through adversarial fine-tuning on numerically perturbed examples. Using parameter-efficient fine-tuning, small Qwen3 models (0.6B$\unicode{x2013}$8B) reach 98.7% accuracy on label-flipping perturbations, outperforming larger zero-shot models and frontier systems (GPT-5.4 Pro (74.0%) and Gemini 2.5 Flash (73.9%)). The gains generalise to unseen perturbation types, indicating robust numerical decision boundaries rather than memorised edits. Robustness also transfers without target-domain data, significantly improving cross-lingual performance in Spanish. We further show that the same fine-tuning recipe confers robustness to evidence-side perturbations, using the VitaminC dataset.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Aarnes, P. R., & Setty, V. (2026). Towards Robust Numerical Claim Verification. https://omanscience.com/en/articles/towards-robust-numerical-claim-verification
MLA 9
Aarnes, Peter Røysland, and Vinay Setty. "Towards Robust Numerical Claim Verification." https://omanscience.com/en/articles/towards-robust-numerical-claim-verification.
Chicago (author–date)
Aarnes, Peter Røysland, and Vinay Setty. 2026. "Towards Robust Numerical Claim Verification." https://omanscience.com/en/articles/towards-robust-numerical-claim-verification.
Harvard
Aarnes, P. R. and Setty, V. (2026) 'Towards Robust Numerical Claim Verification', Available at: https://omanscience.com/en/articles/towards-robust-numerical-claim-verification.
Vancouver
Aarnes PR, Setty V. Towards Robust Numerical Claim Verification. https://omanscience.com/en/articles/towards-robust-numerical-claim-verification
IEEE
P. R. Aarnes, and V. Setty, "Towards Robust Numerical Claim Verification," https://omanscience.com/en/articles/towards-robust-numerical-claim-verification.