Hate speech classification and explainability on HateXplain (test)
0.3176IOU F1Mistral-24.07
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Mistral-24.07Model=Mistral-24.072026.01 | 0.3176 | — | — | — | — | — | — | 0.3934 | — | — | — | |
| GPT-4oModel=GPT-4o2026.01 | 0.3097 | — | — | — | — | — | — | 0.4613 | — | — | — | |
| X-MuTeST with LLMFramework=X-MuTeST, Consultant=LLM2026.01 | 0.3086 | — | — | — | — | — | — | 0.5125 | — | 0.8433 | 0.0507 | |
| Llama-3.1Model=Llama-3.12026.01 | 0.2819 | — | — | — | — | — | — | 0.3995 | — | — | — | |
| XLMR-Rationale-LIMEBackbone=XLM-R, Explainer=LIME, With-Rationale=true2026.01 | 0.2389 | — | — | — | — | — | — | 0.4665 | — | 0.7449 | 0.0559 | |
| BERT-HateXplain-LIMEBackbone=BERT, Training-Source=HateXplain, Explainer=LIME2026.01 | 0.2382 | — | — | — | — | — | — | 0.4454 | — | 0.8555 | 0.2019 | |
| XLMR-Rationale-XMuTeSTBackbone=XLM-R, Explainer=X-MuTeST, With-Rationale=true2026.01 | 0.2357 | — | — | — | — | — | — | 0.4752 | — | 0.7682 | 0.0546 | |
| BiAtt-BiRNN-HateXplainToken Method=Attn2026.01 | 0.23 | 0.65 | 0.64 | 0.81 | 0.734 | 0.724 | 0.706 | 0.487 | 0.823 | 0.31 | -0.06 | |
| BERT-ILPBackbone=BERT, Explainer=ILP2026.01 | 0.2291 | — | — | — | — | — | — | 0.4387 | — | 0.7824 | 0.2008 | |
| XLMR-LIMEBackbone=XLM-R, Explainer=LIME2026.01 | 0.2279 | — | — | — | — | — | — | 0.3886 | — | 0.65 | 0.1244 | |
| BiRNN-HateXplainToken Method=Attn2026.01 | 0.222 | 0.629 | 0.629 | 0.805 | 0.691 | 0.636 | 0.674 | 0.506 | 0.841 | 0.281 | 0.039 | |
| XLMR-XMuTeSTBackbone=XLM-R, Explainer=X-MuTeST2026.01 | 0.2195 | — | — | — | — | — | — | 0.4087 | — | 0.6842 | 0.0945 | |
| Muril-Rationale-XMuTeSTBackbone=MuRIL, Explainer=X-MuTeST, With-Rationale=true2026.01 | 0.2195 | — | — | — | — | — | — | 0.4743 | — | 0.8372 | 0.0795 | |
| Muril-XMuTeSTBackbone=MuRIL, Explainer=X-MuTeST2026.01 | 0.2184 | — | — | — | — | — | — | 0.3928 | — | 0.7527 | 0.0871 | |
| Muril-Rationale-LIMEBackbone=MuRIL, Explainer=LIME, With-Rationale=true2026.01 | 0.2121 | — | — | — | — | — | — | 0.4553 | — | 0.7992 | 0.0856 | |
| Muril-LIMEBackbone=MuRIL, Explainer=LIME2026.01 | 0.2002 | — | — | — | — | — | — | 0.3908 | — | 0.7365 | 0.0898 | |
| BiAtt-BiRNN-HateXplainToken Method=LIME2026.01 | 0.188 | 0.65 | 0.64 | 0.81 | 0.734 | 0.724 | 0.706 | 0.409 | 0.738 | 0.318 | -0.126 | |
| BiRNN-HateXplainToken Method=LIME2026.01 | 0.174 | 0.629 | 0.629 | 0.805 | 0.691 | 0.636 | 0.674 | 0.407 | 0.685 | 0.343 | -0.075 | |
| CNN-GRUToken Method=LIME2026.01 | 0.167 | 0.627 | 0.606 | 0.793 | 0.654 | 0.623 | 0.659 | 0.385 | 0.648 | 0.316 | -0.082 | |
| BiRNN-AttnToken Method=Attn2026.01 | 0.167 | 0.621 | 0.614 | 0.795 | 0.653 | 0.662 | 0.668 | 0.369 | 0.643 | 0.278 | 0.001 | |
| BiRNNToken Method=LIME2026.01 | 0.162 | 0.595 | 0.575 | 0.767 | 0.64 | 0.604 | 0.671 | 0.361 | 0.605 | 0.421 | -0.051 | |
| BiRNN-AttnToken Method=LIME2026.01 | 0.162 | 0.621 | 0.614 | 0.795 | 0.653 | 0.662 | 0.668 | 0.386 | 0.65 | 0.308 | -0.075 |