arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.04102cs.CLcs.AI
收藏

作为事实标注的语言模型?

Language Models as Fact Checkers?

Hong Kong University of Science and Technology(香港科技大学);Facebook AI(Facebook人工智能)

Nayeon Lee, Belinda Z. Li, Sinong Wang, Wen-tau Yih, Hao Ma, Madian Khabsa

摘要:

最近的工作表明,语言模型(LM)存储从预训练数据中学习到的常识和事实知识。在本文中,我们利用这种隐含的知识,创建一个有效的端到端的事实检查器,只使用一个语言模型,没有任何外部知识或显式检索组件。虽然以前的工作中提取知识的LM集中在开放领域的问答的任务,尽我们所知,这是第一次工作,研究使用语言模型作为事实检查。在一个封闭的书设置,我们表明,我们的zero-shot LM方法优于随机基线的标准发热任务,我们微调LM相比,有利的标准基线。虽然我们并没有最终优于使用显式知识库的方法,我们相信我们的探索表明,这种方法是可行的,有很大的探索空间。

英文摘要:

Recent work has suggested that language models (LMs) store both common-sense and factual knowledge learned from pre-training data. In this paper, we leverage this implicit knowledge to create an effective end-to-end fact checker using a solely a language model, without any external knowledge or explicit retrieval components. While previous work on extracting knowledge from LMs have focused on the task of open-domain question answering, to the best of our knowledge, this is the first work to examine the use of language models as fact checkers. In a closed-book setting, we show that our zero-shot LM approach outperforms a random baseline on the standard FEVER task, and that our fine-tuned LM compares favorably with standard baselines. Though we do not ultimately outperform methods which use explicit knowledge bases, we believe our exploration shows that this method is viable and has much room for exploration.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑