Leveraging Speech Acts for Low-Data and Cross-Domain Conversation Derailment Forecasting

arXiv:2608.25359v1 Announce Type: new Abstract: Conversational derailment forecasting aims to predict when online discussions will escalate into hostility, enabling proactive moderation. Existing approaches often struggle in low-data settings and to generalize across domains. This poses a challenge for new platforms and smaller communities where annotated data is limited. We propose modeling pragmatic representations of conversations to reduce lexical noise and improve generalizability. Specifically, speech act information is used as an auxiliary learning signal alongside textual semantics. Experimental results show improved performance across three datasets, particularly in low-data and cross-domain settings.

The Enforcement and Feasibility of Hate Speech Moderation on Twitter
The Cross-Domain Generalization Cost of Offensive Language Detection
When Does Demographic Information Help? Data and Modeling Regimes for Perspective-Aware Hate Speech Detection
AtmosERC: Modeling Dialogue-Level Affective Atmosphere for Emotion Recognition in Conversation
Mapping Election Toxicity on Social Media across Issue, Ideology, and Psychosocial Dimensions

Source: arXiv cs.CL | 2026-08-27

Related