Scholarlink

The paper gives a brief overview of the four shared tasks organized at the PAN 2025 lab on digital text forensics and stylometry to be hosted at CLEF 2025. The goal of the PAN lab is to advance the state-of-the-art in text forensics and stylometry through an objective evaluation of new and established methods on new benchmark datasets. Our three tasks in 2025 will be: (1) generative AI detection, particularly in mixed and obfuscated authorship scenarios, (2) multilingual text detoxification, a continued task that aims re-formulate text in a non-toxic way for multiple languages, and (3) multi-author writing style analysis, a continued task that aims to find positions of authorship change., and(4) generative plagiarism detection, a new task that targets source retrieval and text alignment between generated text and source documents. As with the previous editions, PAN invites software submissions as easy-to-reproduce docker containers; more than 400 pieces of software have been submitted from PAN’12 through PAN’24 combined, with all recent evaluations running on the TIRA experimentation platform [11].

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Overview of PAN 2025: Generative AI Detection, Multilingual Text Detoxification, Multi-author Writing Style Analysis, and Generative Plagiarism Detection

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Janek Bevendorff,
  • Daryna Dementieva,
  • Maik Fröbe,
  • Bela Gipp,
  • André Greiner-Petter,
  • Jussi Karlgren,
  • Maximilian Mayerl,
  • Preslav Nakov,
  • Alexander Panchenko,
  • Martin Potthast,
  • Artem Shelmanov,
  • Efstathios Stamatatos,
  • Benno Stein,
  • Yuxia Wang,
  • Matti Wiegmann,
  • Eva Zangerle

摘要

The paper gives a brief overview of the four shared tasks organized at the PAN 2025 lab on digital text forensics and stylometry to be hosted at CLEF 2025. The goal of the PAN lab is to advance the state-of-the-art in text forensics and stylometry through an objective evaluation of new and established methods on new benchmark datasets. Our three tasks in 2025 will be: (1) generative AI detection, particularly in mixed and obfuscated authorship scenarios, (2) multilingual text detoxification, a continued task that aims re-formulate text in a non-toxic way for multiple languages, and (3) multi-author writing style analysis, a continued task that aims to find positions of authorship change., and(4) generative plagiarism detection, a new task that targets source retrieval and text alignment between generated text and source documents. As with the previous editions, PAN invites software submissions as easy-to-reproduce docker containers; more than 400 pieces of software have been submitted from PAN’12 through PAN’24 combined, with all recent evaluations running on the TIRA experimentation platform [11].