README / README.md
dardem's picture
Update README.md
a0d9085 verified
|
raw
history blame
703 Bytes
metadata
title: README
emoji: πŸ“Š
colorFrom: green
colorTo: indigo
sdk: static
pinned: false

Demarked: A Strategy for Enhanced Abusive Speech Moderation through Counterspeech, Detoxification, and Message Management

Proactive abusive textual speech moderation pipelines with NLP:

  • abusive speech classification;
  • text detoxification;
  • counter speech generation.

Position paper: Yimam, S.M., Dementieva, D., Fischer, T., Moskovskiy, D., Rizwan, N., Saha, P., Roy, S., Semmann, M., Panchenko, A., Biemann, C. and Mukherjee, A., 2024. Demarked: A Strategy for Enhanced Abusive Speech Moderation through Counterspeech, Detoxification, and Message Management. https://arxiv.org/abs/2406.19543