提出无预设的问答分解框架，提升大模型验证声明的稳定性与准确性。
If We May De-Presuppose: Robustly Verifying Claims through Presupposition-Free Question Decomposition
- 通过去预设的问答分解机制，避免生成含未验证假设的问题。
- 在多个数据集和模型上，性能提升2-5%，显著降低提示敏感性。
- 适合需要高可信度推理的场景，如事实核查与可信AI系统。
已有研究指出，生成问题中的预设会引入未经验证的假设，导致声明验证不一致。同时，大语言模型（LLMs）的提示敏感性仍是重大挑战，性能波动可达3-6%。尽管近期进展有所缓解，我们的研究证实提示敏感性仍普遍存在。为此，我们提出一种结构化且鲁棒的声明验证框架，通过无预设的分解问题进行推理。在多种提示、数据集和模型上的大量实验表明，即使最先进的模型也易受提示变化影响且存在预设问题。所提方法能持续缓解这些问题，性能最高提升2-5%。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。