If We May De-Presuppose: Robustly Verifying Claims through Presupposition-Free Question Decomposition

Shubhashis Roy Dipta, Francis Ferraro


Abstract
Prior work has shown that presupposition in generated questions can introduce unverified assumptions, leading to inconsistencies in claim verification. Additionally, prompt sensitivity remains a significant challenge for large language models (LLMs), resulting in performance variance as high as **3–6%**. While recent advancements have reduced this gap, our study demonstrates that prompt sensitivity remains a persistent issue. To address this, we propose a structured and robust claim verification framework that reasons through presupposition-free, decomposed questions. Extensive experiments across multiple prompts, datasets, and LLMs reveal that even state-of-the-art models remain susceptible to prompt variance and presupposition. Our method consistently mitigates these issues, achieving up to a **2–5%** improvement.
Anthology ID:
2025.starsem-1.20
Volume:
Proceedings of the 14th Joint Conference on Lexical and Computational Semantics (*SEM 2025)
Month:
November
Year:
2025
Address:
Suzhou, China
Editors:
Lea Frermann, Mark Stevenson
Venue:
*SEM
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
253–266
Language:
URL:
https://preview.aclanthology.org/ingest-emnlp/2025.starsem-1.20/
DOI:
Bibkey:
Cite (ACL):
Shubhashis Roy Dipta and Francis Ferraro. 2025. If We May De-Presuppose: Robustly Verifying Claims through Presupposition-Free Question Decomposition. In Proceedings of the 14th Joint Conference on Lexical and Computational Semantics (*SEM 2025), pages 253–266, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
If We May De-Presuppose: Robustly Verifying Claims through Presupposition-Free Question Decomposition (Roy Dipta & Ferraro, *SEM 2025)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-emnlp/2025.starsem-1.20.pdf