Automated soft moderation systems are unable to ascertain if a post supports or refutes a false claim, resulting in a large number of contextual false positives. This limits their effectiveness, for example undermining trust in health experts by adding warnings to their posts or resorting to vague warnings instead of granular fact-checks, which result in desensitizing users. In this paper, we propose to incorporate stance detection into existing automated soft-moderation pipelines, with the goal of ruling out contextual false positives and providing more precise recommendations for social media content that should receive warnings. We develop a textual deviation task called Contrastive Textual Deviation (CTD) and show that it outperforms existing stance detection approaches when applied to soft moderation.We then integrate CTD into the stateof-the-art system for automated soft moderation Lambretta, showing that our approach can reduce contextual false positives from 20% to 2.1%, providing another important building block towards deploying reliable automated soft moderation tools on social media.
Lightweight Stylistic Consistency Profiling: Robust Detection of LLM-Generated Textual Content for Multimedia Moderation
School of Computer Science, Shanghai Jiao Tong University(上海交通大学计算机科学学院);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);College of Computer Science and Technology, Zhejiang University(浙江大学计算机科学与技术学院)
CoT-MoTE: Exploring ConTextual Masked Auto-Encoder Pre-training with Mixture-of-Textual-Experts for Passage Retrieval
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院);Kuaishou Technology(快手科技)
Probing Association Biases in LLM Moderation Over-Sensitivity
Department of Computer Science, Dartmouth College(达特茅斯大学计算机科学系);Department of Computer Science and Engineering, The Ohio State University(俄亥俄州立大学计算机科学与工程系)
本文研究LLM在内容审查中因主题关联偏差导致的过灵敏问题,提出Topic Association Analysis方法,发现高级模型在误报中存在更强的主题关联偏移,表明需结合关键词过滤与学习主题关联来缓解。
Conversational Speech Recognition by Learning Audio-textual Cross-modal Contextual Representation
School of Computer Science, Northwestern Polytechnical University(西北工业大学计算机学院);School of Computer Science and Engineering, Northeastern University(东北大学计算机科学与工程学院);Mashang Consumer Finance Co., Ltd.(马上消费金融股份有限公司)
Dialogizer: Context-aware Conversational-QA Dataset Generation from Textual Sources
LG AI Research(LG AI研究院);IPAI, Seoul National University(首尔大学IPAI);Dept. of ECE, Seoul National University(首尔大学电子与计算机工程系);Chung-Ang University(中央大学);SNU-LG AI Research Center(首尔大学-LG AI研究中心)
The Need for an External Observer Formalizing the Sufficiency Gap: A Mathematical Extension of Mixture Identifiability and Contextual Grounding in Sequence Models