Binary Hate Detection on H-VLI Easy
96.94AccuracyQwen-VL-Plus
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen-VL-PlusTraining=False2026.03 | 96.94 | |
| Qwen-VL-MaxTraining=False2026.03 | 95 | |
| GPT-5-miniTraining=False2026.03 | 94.86 | |
| GPT-4oTraining=False2026.03 | 94.72 | |
| GPT-5-miniTraining=False, Framework=ARCADE2026.03 | 94.44 | |
| Qwen3-VL-PlusTraining=False2026.03 | 94.29 | |
| GPT-5Training=False2026.03 | 93.59 | |
| Gemini-2.5-FlashTraining=False2026.03 | 93.19 | |
| Qwen3-VL-PlusTraining=False, Framework=ARCADE2026.03 | 92.76 | |
| GPT-5Training=False, Framework=ARCADE2026.03 | 92.58 | |
| GPT-4.1-miniTraining=False2026.03 | 91.67 | |
| GPT-4.1-miniTraining=False, Framework=ARCADE2026.03 | 91.52 | |
| Gemini-2.5-ProTraining=False2026.03 | 91.11 | |
| Qwen-VL-MaxTraining=False, Framework=ARCADE2026.03 | 89.69 | |
| GPT-4oTraining=False, Framework=ARCADE2026.03 | 89.57 | |
| Gemini-2.5-ProTraining=False, Framework=ARCADE2026.03 | 87.48 | |
| Gemini-2.5-FlashTraining=False, Framework=ARCADE2026.03 | 85.95 | |
| Qwen-VL-PlusTraining=False, Framework=ARCADE2026.03 | 83.66 | |
| PromptHateTraining=True2026.03 | 76.67 | |
| BERT+ViTTraining=True2026.03 | 75.83 | |
| MMBTTraining=True2026.03 | 74.44 | |
| BERTTraining=True2026.03 | 73.33 | |
| MomentaTraining=True2026.03 | 70.42 | |
| ViT-b-16Training=True2026.03 | 48.89 |