根据事实声明的风险等级动态调整验证强度，提升长文本生成准确性
Not All Claims Are Equally Risky: FACTOR for Adaptive Verification in Factual Long-Form Generation
- 按每条声明的不确定性动态分配验证资源
- 在FactScore上同时提升事实准确率并降低验证开销
- 适用于各类大模型，无需修改训练过程
大语言模型生成流畅的长篇文本，但常引入无依据的事实性陈述。现有验证方法通过外部证据约束生成，但对所有声明采用相同验证策略，忽视了不同声明的幻觉风险差异。本文提出FACTOR（FACTuality-Oriented Risk-aware Verification），一种推理阶段的自适应验证框架，根据声明级别的不确定性动态调整验证标准。FACTOR结合不确定性估计、自适应语言推理验证和候选重排序，将验证资源集中于高风险声明。在FactScore基准上的实验表明，自适应验证在提升事实准确性的同时显著降低验证成本。消融研究进一步揭示了性能提升的主要驱动因素。结果证明FACTOR在提升长文本生成事实性方面具有高效且模型无关的特性。
原文摘要 · Abstract (English)
Large Language Models (LLMs) generate fluent long-form text, however, often add unsupported factual claims. Existing verification techniques improve factuality by grounding generation in external evidence. However, the same verification policy usually applies to all claims despite being differences in hallucination risks. We propose \textit{FACTOR} (\textit{FACTuality-Oriented Risk-aware Verification}), an inference-time model that adapts verification criteria according to claim-level uncertainty. FACTOR combines uncertainty estimation, adaptive language inference verification, and candidate re-ranking to allocate verification effort where it is most needed. We evaluate \textit{FACTOR} on FactScore benchmark showing that adaptive verification improves factuality while reducing verification cost simultaneously. We further perform different ablation studies to identify the primary driver of these gains. Our results show the effective and model-agnostic performance of \textit{FACTOR} for improving factuality in long-form generation.
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。