Classification and Explainability on Hindi
87.45AccuracyX-MuTeST with LLM
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| X-MuTeST with LLMFramework=X-MuTeST, Backbone=LLM2026.01 | 87.45 | 81.68 | 86.23 | 43.44 | 26.67 | 74.83 | 47.68 | |
| XLMR-Rationale-LIMEBackbone=XLM-R, Rationale Training=true, Explainability Method=LIME2026.01 | 85.9 | 80 | 84.56 | 31.27 | 15.64 | 70.11 | 53.14 | |
| XLMR-Rationale-XMuTeSTBackbone=XLM-R, Rationale Training=true, Explainability Method=X-MuTeST2026.01 | 85.9 | 80 | 84.56 | 40.07 | 18.32 | 72.61 | 56.12 | |
| BERT-HateXplain-LIMEBackbone=BERT-HateXplain, Explainability Method=LIME2026.01 | 85.57 | 78.84 | 83.76 | 28.22 | 14.7 | 64.81 | 21.11 | |
| XLMR-LIMEBackbone=XLM-R, Explainability Method=LIME2026.01 | 85.35 | 79.88 | 84.18 | 30.68 | 14.94 | 66.51 | 72.96 | |
| XLMR-XMuTeSTBackbone=XLM-R, Explainability Method=X-MuTeST2026.01 | 85.35 | 79.88 | 84.18 | 39.48 | 17.99 | 71.22 | 65.28 | |
| BERT-ILPBackbone=BERT, Explainability Method=ILP2026.01 | 83.95 | 79.18 | 83.02 | 39.41 | 26.18 | 62.99 | 27.04 | |
| Muril-Rationale-LIMEBackbone=MuRIL, Rationale Training=true, Explainability Method=LIME2026.01 | 83.57 | 77.16 | 82.17 | 31.6 | 17.24 | 68.9 | 49.36 | |
| Muril-Rationale-XMuTeSTBackbone=MuRIL, Rationale Training=true, Explainability Method=X-MuTeST2026.01 | 83.57 | 77.16 | 82.17 | 41.17 | 19.06 | 69.45 | 50.42 | |
| Muril-LIMEBackbone=MuRIL, Explainability Method=LIME2026.01 | 82.69 | 76.79 | 81.49 | 31.24 | 15.88 | 55.85 | 55.17 | |
| Muril-XMuTeSTBackbone=MuRIL, Explainability Method=X-MuTeST2026.01 | 82.69 | 76.79 | 81.49 | 41.82 | 17.34 | 54.86 | 56.15 | |
| Mistral-24.07Model Type=LLM2026.01 | 71.59 | 66.84 | 70.99 | 39.75 | 28.32 | — | — | |
| GPT-4oModel Type=LLM2026.01 | 54.61 | 60.71 | 55.42 | 39.42 | 22.66 | — | — | |
| Llama-3.1Model Type=LLM2026.01 | 51.72 | 58.29 | 50.49 | 42.63 | 30.63 | — | — |