aclanthology.org web signal

EMNLP tutorial maps open challenges in hate-speech detection

By Published Updated

TL;DR

  • The EMNLP 2024 tutorial in Miami, by five researchers, treats online hate-speech moderation as a set of open challenges, not a solved product.
  • Topics include multilingualism, counter-narrative generation, a hands-on session with a popular hate-speech detection API, and fairness and ethics in AI.
  • The abstract names no specific API, no detector numbers, and frames safer online spaces as a collaboration goal rather than a shipped technology.

EMNLP 2024's tutorial program in Miami featured 'Countering Hateful and Offensive Speech Online - Open Challenges,' a session that frames online hate-speech moderation as a field still full of unresolved problems rather than a shipped product.

The abstract on the ACL Anthology lays out what the five authors, Flor Miriam Plaza-del-Arco, Debora Nozza, Marco Guerini, Jeffrey Sorensen, and Marcos Zampieri, intend to cover: 'multilingualism, counter-narrative generation, a hands-on session with one of the most popular APIs for detecting hate speech, fairness, and ethics in AI, and the use of recent advanced approaches.' The abstract names neither the API nor any specific detector numbers; the link reached us via two researchers we follow in the field.

The authors position the session as applied rather than purely academic, aiming to 'foster collaboration and inspire participants to create safer online spaces by detecting and mitigating hate speech.'

Shared on Bluesky by 2 AI experts

Originally reported by aclanthology.org

Read the original article →

Original headline: Countering Hateful and Offensive Speech Online - Open Challenges