用概率推理评估大模型长文本事实性，提升生成内容可信度。
FactReasoner: A Probabilistic Approach to Long-Form Factuality Assessment for Large Language Models
- 将长文本分解为原子单元，结合外部知识进行逻辑关系建模。
- 在标注和未标注数据集上，事实精确率与召回率均优于现有方法。
- 适合需要高可信输出的场景，如医疗、法律等专业领域应用。
大型语言模型（LLMs）在生成任务中表现卓越，但其输出常缺乏事实准确性，限制了在对正确性要求高的实际应用中的可靠性。本文提出FactReasoner，一种基于神经符号的新型事实性评估框架，采用概率推理来评估长文本生成内容的真实性。该框架将生成文本分解为原子单元，从外部知识源检索相关上下文信息，并利用概率编码建模这些单元与其上下文之间的逻辑关系（如蕴含、矛盾）。随后，估计每个原子单元由检索证据支持的后验概率。在标注与未标注基准数据集上的实验表明，FactReasoner在事实精确率和召回率方面通常优于现有的基于提示的方法。开源代码已公开：https://github.com/IBM/FactReasoner。
原文摘要 · Abstract (English)
Large language models (LLMs) have achieved remarkable success in generative tasks, yet they often fall short in ensuring the factual accuracy of their outputs, thus limiting their reliability in real-world applications where correctness is critical. In this paper, we present FactReasoner, a novel neuro-symbolic based factuality assessment framework that employs probabilistic reasoning to evaluate the truthfulness of long-form generated responses. FactReasoner decomposes a response into atomic units, retrieves relevant contextual information from external knowledge sources, and models the logical relationships (e.g., entailment, contradiction) between these units and their contexts using probabilistic encodings. It then estimates the posterior probability that each atomic unit is supported by the retrieved evidence. Our experiments on both labeled and unlabeled benchmark datasets demonstrate that FactReasoner often outperforms state-of-the-art prompt-based methods in terms of factual precision and recall. Our open-source implementation is publicly available at: https://github.com/IBM/FactReasoner.
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。