arXiv 2602.095162026-02-10cs.CL

CLEF-2026 CheckThat! 实验室：推进多语言事实核查

The CLEF-2026 CheckThat! Lab: Advancing Multilingual Fact-Checking

University of Applied Sciences Potsdam, Germany（应用科学大学波茨坦，德国）；GESIS - Leibniz Institute for the Social Sciences, Cologne, Germany（GESIS—社会科学院莱比锡研究所，德国）；Stockholm University, Sweden（斯德哥尔摩大学，瑞典）；University of Stavanger, Norway（斯塔万格大学，挪威）；Indian Institute of Technology Delhi, India（印度理工学院德里，印度）；Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi（穆罕默德·本·扎耶德人工智能大学，阿布扎比）；Delft University of Technology, The Netherlands（代尔夫特理工大学，荷兰）；Independent Researcher（独立研究者）；médialab Sciences Po, Paris, France（Sciences Po 媒介实验室，法国）；University of Montpellier, LIRMM, CNRS, Montpellier, France（蒙彼利埃大学，LIRMM，CNRS，法国）

Julia Maria Struß, Sebastian Schellhammer, Stefan Dietze, Venktesh V, Vinay Setty, Tanmoy Chakraborty, Preslav Nakov, Avishek Anand, Primakov Chungkham, Salim H… 展开作者

Julia Maria Struß, Sebastian Schellhammer, Stefan Dietze, Venktesh V, Vinay Setty, Tanmoy Chakraborty, Preslav Nakov, Avishek Anand, Primakov Chungkham, Salim Hafid, Dhruv Sahnan, Konstantin Todorov

备注：: misinformation, disinformation, fact-checking, claim source retrieval, generating fact-checking articles

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

CLEF-2026 CheckThat! 实验室通过多语言事实核查任务推进验证流程的发展，涵盖来源检索、数字时间声明核查及生成完整核查文章等挑战性任务。

摘要：

CheckThat! 实验室旨在推进对抗在线通信中虚假信息和操纵行为的创新技术发展，涵盖多种语言和平台。虽然早期版本聚焦于验证流程的核心任务（可信度检查、证据检索和验证），但在过去三年中，实验室增加了与验证过程相关的额外任务。在本年度版本中，验证流程再次成为核心，包含三个任务：任务1是针对科学网络声明的来源检索（2025版的延续），任务2是针对数字和时间声明的事实核查，增加了2025版的推理组件，任务3则通过生成完整事实核查文章扩展验证流程。这些任务代表了具有挑战性的分类和检索问题，以及文档和跨度层面的生成挑战，包括多语言环境。

英文摘要：

The CheckThat! lab aims to advance the development of innovative technologies combating disinformation and manipulation efforts in online communication across a multitude of languages and platforms. While in early editions the focus has been on core tasks of the verification pipeline (check-worthiness, evidence retrieval, and verification), in the past three editions, the lab added additional tasks linked to the verification process. In this year's edition, the verification pipeline is at the center again with the following tasks: Task 1 on source retrieval for scientific web claims (a follow-up of the 2025 edition), Task 2 on fact-checking numerical and temporal claims, which adds a reasoning component to the 2025 edition, and Task 3, which expands the verification pipeline with generation of full-fact-checking articles. These tasks represent challenging classification and retrieval problems as well as generation challenges at the document and span level, including multilingual settings.

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。