| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| CM, EN, and HI (test) | Hing-RoBERTa Trilingual | Consistency0.6928 | 42 | 2mo ago | |
| Hindi (HI) (test) | Hing-RoBERTa Mixed | Macro F171.96 | 42 | 2mo ago | |
| English (EN) (test) | XLM-R Trilingual | Macro F1 Score70.03 | 42 | 2mo ago | |
| Code-mixed (CM) (test) | XLM-R Trilingual | Macro F169.72 | 42 | 2mo ago | |
| ToxiGen (test) | DSA:LST | AUC99 | 24 | 1mo ago | |
| HateXplain | CREDENCE-DeBERTa | Accuracy0.763 | 21 | 1mo ago | |
| Stormfront (test) | xList-Hate | AUC0.921 | 16 | 3mo ago | |
| Measuring Hate Speech (MHS) (test) | xList-Hate | AUC0.819 | 16 | 3mo ago | |
| HateCheck (test) | xList-Hate | AUC95.6 | 16 | 3mo ago | |
| ETHOS (test) | xList-Hate | AUC0.912 | 16 | 3mo ago | |
| MHS small models | F1 Score61.99 | 14 | 7d ago | ||
| MHS | F1 Score70.6 | 14 | 7d ago | ||
| Implicit Hate (test) | Llama_Full | Macro F10.67 | 13 | 1mo ago | |
| Latent Hate (test) | SMARTER (Llama_DPO-256) | Macro F1 Score69 | 13 | 1mo ago | |
| HateXplain (test) | Llama_Full | Macro-F172 | 13 | 1mo ago | |
| ETHOS binary | BILSTM | F1 Score79.71 | 12 | 3mo ago | |
| Internal Set | Precision97 | 8 | 3mo ago | ||
| Hate Speech | Precision75.9 | 6 | 1mo ago | ||
| Ethos Benchmark | Precision91 | 5 | 3mo ago | ||
| HateXplain standard (test) | AdapLeR | Accuracy68.6 | 5 | 3mo ago | |
| HTweet-MC | AT1-unk | AUC0.938 | 4 | 3mo ago |
