Detecting Community Sensitive Norm Violations in Online Conversations
University of Michigan(密歇根大学);Paul G. Allen School of Computer Science & Engineering, University of Washington(华盛顿大学保罗·G·艾伦计算机科学与工程学院);Language Technologies Institute Carnegie Mellon University(卡内基梅隆大学语言技术研究所)
Chan Young Park, Julia Mendelsohn, Karthik Radhakrishnan, Kinjal Jain, Tushar Kanakagiri, David Jurgens, Yulia Tsvetkov
Online platforms and communities establish their own norms that govern what behavior is acceptable within the community. Substantial effort in NLP has focused on identifying unacceptable behaviors and, recently, on forecasting them before they occur. However, these efforts have largely focused on toxicity as the sole form of community norm violation. Such focus has overlooked the much larger set of rules that moderators enforce. Here, we introduce a new dataset focusing on a more complete spectrum of community norms and their violations in the local conversational and global community contexts. We introduce a series of models that use this data to develop context- and community-sensitive norm violation detection, showing that these changes give high performance.
Detecting Speaker Personas from Conversational Texts
Microsoft Research Asia(微软亚洲研究院);State Key Laboratory of Cognitive Intelligence, iFLYTEK Research(科大讯飞研究院认知智能国家重点实验室)
人物角色对对话响应预测很有用。然而,目前的研究中使用的人物角色是预先定义的,很难在对话之前获得。为了解决这个问题,我们研究了一个新的任务,命名为扬声器人物角色检测(SPD),其目的是检测扬声器人物角色的基础上,普通的会话文本。在这个任务中,一个最匹配的人物角色是从给定的会话文本的候选人。这是一个多对多的语义匹配任务,因为SPD中的上下文和人物角色都由多个句子组成。这些句子之间的长期依存和动态冗余增加了这项任务的难度。我们为SPD构建了一个数据集,称为Persona Match on Persona Chat(PMPC)。此外,我们评估了几个基线模型,并提出了话语轮廓(U2 P)匹配网络这项任务。U2 P模型以细粒度操作,将上下文和人物角色视为多个序列的集合。然后,对每个序列对进行评分,并通过聚合获得上下文-人物角色对的可解释的总得分。评估结果表明,U2 P模型的性能明显优于基准模型。
Attention-based Modeling for Emotion Detection and Classification in Textual Conversations
LIRMM UMR 5506;CNRS(法国国家科学研究中心);University of Montpellier(蒙彼利埃大学);IUT de Béziers(贝济耶大学技术学院);AMIS;Paul Valery University - Montpellier 3(保罗·瓦莱里大学-蒙彼利埃第三大学)
Overview of the PsyDefDetect Shared Task at BioNLP 2026: Detecting Levels of Psychological Defense Mechanisms in Supportive Conversations
University of Technology Sydney(技术大学悉尼);Xi’an Jiaotong-Liverpool University(西安交通大学-利物浦大学);University of Utah(犹他大学);Harvard University(哈佛大学);The University of Melbourne(墨尔本大学);The University of Manchester(曼彻斯特大学);ELLIS Institute Finland(芬兰ELLIS研究所);University of Turku(图尔库大学)
You Never Know a Person, You Only Know Their Defenses: Detecting Levels of Psychological Defense Mechanisms in Supportive Conversations
University of Technology Sydney(悉尼技术大学);Xi’an Jiaotong-Liverpool University(西安交通大学利物浦大学);University of Utah(犹他大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州));Harvard University(哈佛大学);The University of Sydney(悉尼大学);ELLIS Institute Finland(芬兰ELLIS研究所);University of Turku(图尔库大学)
Detecting early signs of depression in the conversational domain: The role of transfer learning in low-resource scenarios
PRHLT research center, Universitat Politècnica de València(瓦伦西亚理工大学PRHLT研究中心);Faculty of Mathematics and Computer Science, University of Bucharest(布加勒斯特大学数学与计算机科学学院);CIIRC, Czech Technical University(捷克技术大学捷克信息学、机器人与控制论研究所)
Beyond Static: Related Questions Retrieval Through Conversations in Community Question Answering
School of Computer Science and Engineering, University of Electronic Science and Technology of China(电子科技大学计算机科学与工程学院);School of Management Science and Engineering, Southwestern University of Finance and Economics(西南财经大学管理科学与工程学院)
Conversations Gone Alright: Quantifying and Predicting Prosocial Outcomes in Online Conversations
Carnegie Mellon University(卡内基梅隆大学);The Hong Kong University of Science and Technology(香港科技大学);University of Michigan(密歇根大学);University of Illinois, Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)