CoGate-LSTM: Prototype-Guided Feature-Space Gating for Mitigating Gradient Dilution in Imbalanced Toxic Comment Classification

Researchers at Nemati AI have developed CoGate-LSTM, a new model that uses cosine-similarity feature gating to mitigate gradient dilution in the classification of imbalanced toxic comments. This architecture outperforms existing models like BERT and XGBoost on macro-F1 scores while using fewer parameters and offering faster inference times, making it particularly effective for detecting rare but critical toxic labels such as severe_toxic and threat.

Read the full article at arXiv cs.CL (NLP)

---

Want to create content about this topic? Use Nemati AI tools to generate articles, social posts, and more.