CLAN

该数据集包含了经过事实核查的社交媒体帖子，以及核查人员在验证过程中创建的简洁、以主张为导向的摘要（标准化主张）。为了防止偏见并确保数据的高质量，该数据集避免了外部注释。同时，它包含了单一和双重参考的标准化主张，以提高鲁棒性。该数据集的规模为6,388个实例，其任务是主张标准化。

This dataset comprises fact-checked social media posts, along with concise, claim-oriented summaries (standardized claims) created by verifiers during the validation process. To avoid bias and ensure high-quality data, this dataset does not use external annotations. Additionally, it includes both single-reference and dual-reference standardized claims to enhance robustness. The dataset has a total of 6,388 instances, and its target task is claim standardization.