If We May De-Presuppose: Robustly Verifying Claims through Presupposition-Free Question Decomposition
This addresses robustness issues in claim verification for LLM applications, though it appears incremental as it builds on prior work to reduce known bottlenecks.
The paper tackles the problem of presupposition in generated questions and prompt sensitivity in large language models (LLMs) for claim verification, proposing a structured framework using presupposition-free question decomposition that achieves up to 2-5% improvement.
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.