Diálogos Tóxicos: Gatilhos e Padrões de Interação no Reddit Brasileiro

2026-04-12 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, short

Summary

A study titled "Diálogos Tóxicos: Gatilhos e Padrões de Interação no Reddit Brasileiro" analyzes the structural and linguistic dynamics of online toxicity within Reddit discussion trees, specifically focusing on how trigger comments escalate conflicts in Brazilian Portuguese. Researchers utilized a fine-tuned BERTAbaporu model to examine these interactions. The findings, presented at the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) in Salvador, Brazil, indicate that toxic discussions are deeper, more engaging, and initially semantically cohesive, but their quality degrades over time. In contrast, non-toxic interactions tend to emphasize social bonding. This research aims to enhance the understanding of toxicity escalation and support the early detection of discursive conflicts.

Key takeaway

For research scientists developing content moderation systems, understanding the dynamics of toxic dialogues is crucial. You should integrate models like BERTAbaporu to analyze discussion depth, engagement, and semantic cohesion over time. This approach can help in proactively identifying trigger comments and predicting the escalation of conflicts, enabling earlier intervention to maintain healthier online environments.

Key insights

Online toxicity in Brazilian Portuguese Reddit discussions escalates from trigger comments, showing initial cohesion but eventual degradation.

Principles

• Toxic discussions are deeper and more engaging.
• Semantic cohesion degrades in toxic interactions.
• Non-toxic interactions foster social bonding.

Method

A fine-tuned BERTAbaporu model was used to analyze structural and linguistic dynamics of online toxicity in Reddit discussion trees.

In practice

• Identify trigger comments in online forums.
• Monitor semantic cohesion degradation.
• Detect early signs of discursive conflicts.

Topics

• Online Toxicity
• Reddit Discussions
• Brazilian Portuguese NLP
• Trigger Comments
• BERTAbaporu Model

Best for: Research Scientist, AI Scientist, NLP Engineer, AI Ethicist

Related on AIssential

• Anatomy of Data Repositories for the Analysis and Detection of Toxicity in Portuguese — Lexical diversity and annotation practices significantly impact cross-domain generalization in hate speech detection datasets.
• Implicit and Indirect: Detecting Face-threatening and Paired Actions in Asynchronous Online Conversations — Detecting face-threatening actions in asynchronous conversations requires handling ambiguity and multiple interpretations.
• Cohesion-6K: An Arabic Dataset for Analyzing Social Cohesion and Conflict in Online Discourse — Divisive online discourse consistently attracts disproportionately higher user engagement compared to unifying narratives.
• PluRule: A Benchmark for Moderating Pluralistic Communities on Social Media — AI models struggle with context-dependent content moderation in pluralistic online communities, performing only slightly better than a trivial baseline.
• Mapping the Political Discourse in the Brazilian Chamber of Deputies: A Multi-Faceted Computational Approach — Computational methods can reveal hidden patterns in parliamentary discourse beyond voting records.
• From Comments to Trends: Studi NLP dan Computational Linguistics dalam Mendeteksi Topik Viral dari… — NLP and computational linguistics are crucial for detecting social media trends from vast, informal, and diverse comment data.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Paper Index on ACL Anthology.