Research map / Large language models / Language, safety and interpretability
Hindi-English Hate Speech Detection: Author Profiling, Debiasing, and Practical Perspectives
Shivang Chopra, Ramit Sawhney, Puneet Mathur, Rajiv Ratn Shah
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add DeepWalk graph embeddings to CNN+BiLSTM+Attn (ablation) | seq2seq CNN+BiLSTM+Attn · HS | accuracy: 0.62 -> 0.71 | Improved |
| use DeepWalk instead of node2vec embeddings (ablation) | seq2seq CNN+BiLSTM+Attn · HS | F1: 0.63 -> 0.67 | Improved |
| add node2vec embeddings (ablation) | seq2seq CNN+BiLSTM+Attn · HS | F1: 0.62 -> 0.63 | Improved |
| add profanity vector on top of DeepWalk (ablation) | seq2seq CNN+BiLSTM+Attn+DW · HS | F1: 0.67 -> 0.73 | Improved |
| add DeepWalk graph embeddings to fine-tuned model (ablation) | FT CNN+BiLSTM+Attn+PV · HS | F1: 0.57 -> 0.64 | Improved |
| add bias elimination (ablation) | FT CNN+BiLSTM+Attn+PV+DW · HS | F1: 0.64 -> 0.73 | Improved |
| add bias elimination (ablation) | FT CNN+BiLSTM+Attn+PV · HS | F1: 0.57 -> 0.64 | Improved |
| add bias elimination (ablation) | FT CNN+BiLSTM+Attn+PV · HEOT | accuracy: 0.82 -> 0.85+ | Improved |
| use CNN+BiLSTM+Attn instead of plain seq2seq+CNN backbone (ablation) | seq2seq text model · HS | F1: 0.49 -> 0.62 | Improved |
| fine-tune word2vec instead of seq2seq translation preprocessing (with PV) | CNN+BiLSTM+Attn+PV · HS | F1: 0.64 -> 0.57 | Hurt |
Closest papers by meaning
- Spanning the Spectrum of Hatred Detection: A Persian Multi-Label Hate Speech Dataset with Annotator Rationales (AAAI 2024 · similarity 0.88)
- IndicCONAN: A Multilingual Dataset for Combating Hate Speech in Indian Context (AAAI 2024 · similarity 0.88)
- A Multi-Task Benchmark for Abusive Language Detection in Low-Resource Settings (NeurIPS 2025 · similarity 0.86)
- Multi-Modal Discussion Transformer: Integrating Text, Images and Graph Transformers to Detect Hate Speech on Social Media (AAAI 2024 · similarity 0.85)
- TRACE: Textual Relevance Augmentation and Contextual Encoding for Multimodal Hate Detection (AAAI 2026 · similarity 0.84)
- BeyondGender: A Multifaceted Bilingual Dataset for Practical Sexism Detection (AAAI 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
