Binary Classification on Apple-defined Content Rating Descriptors
87.62R+ Metric ValueGemini-2.5-Flash
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Gemini-2.5-FlashDescriptor=Unrestricted Web Access2026.05 | 87.62 | 97.89 | |
| QwenSafeDescriptor=Contests2026.05 | 74.36 | 98.62 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Unrestricted Web Access2026.05 | 74.29 | 97.01 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Contests2026.05 | 71.79 | 98.85 | |
| Gemini-2.5-FlashDescriptor=Contests2026.05 | 71.79 | 98.38 | |
| QwenSafeDescriptor=Unrestricted Web Access2026.05 | 60.95 | 93.84 | |
| QwenSafeDescriptor=Realistic Violence2026.05 | 60.27 | 96.64 | |
| QwenSafeDescriptor=Medical / Treatment Information2026.05 | 57.66 | 94.33 | |
| Gemini-2.5-FlashDescriptor=Cartoon or Fantasy Violence2026.05 | 55.97 | 88.69 | |
| Gemini-2.5-FlashDescriptor=Medical / Treatment Information2026.05 | 55.86 | 92.41 | |
| QwenSafeDescriptor=Mature or Suggestive Themes2026.05 | 55.48 | 79.78 | |
| QwenSafeDescriptor=Alcohol, Tobacco or Drug Use or References2026.05 | 54.22 | 91.15 | |
| Gemini-2.5-FlashDescriptor=Mature or Suggestive Themes2026.05 | 52.3 | 75.89 | |
| QwenSafeDescriptor=Cartoon or Fantasy Violence2026.05 | 52.2 | 91.1 | |
| Gemini-2.5-FlashDescriptor=Realistic Violence2026.05 | 52.05 | 94.78 | |
| QwenSafeDescriptor=Sexual Content or Nudity2026.05 | 51.45 | 92.29 | |
| Qwen3-VL-8BDescriptor=Contests2026.05 | 51.28 | 98.02 | |
| QwenSafeDescriptor=Average2026.05 | 51.05 | 92.63 | |
| Gemini-2.5-FlashDescriptor=Average2026.05 | 50.01 | 90.34 | |
| Gemini-2.5-FlashDescriptor=Horror or Fear Themes2026.05 | 48.7 | 90.95 | |
| QwenSafeDescriptor=Profanity or Crude Humour2026.05 | 46.28 | 88.55 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Horror or Fear Themes2026.05 | 45.22 | 93.22 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Medical / Treatment Information2026.05 | 45.05 | 93.45 | |
| Gemini-2.5-FlashDescriptor=Alcohol, Tobacco or Drug Use or References2026.05 | 43.37 | 85.22 | |
| Gemini-2.5-FlashDescriptor=Sexual Content or Nudity2026.05 | 42.75 | 87.83 | |
| Gemini-2.5-FlashDescriptor=Profanity or Crude Humour2026.05 | 41.49 | 82.54 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Simulated Gambling2026.05 | 40.54 | 97.72 | |
| QwenSafeDescriptor=Horror or Fear Themes2026.05 | 40 | 92.38 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Average2026.05 | 37.52 | 91.15 | |
| Qwen3-VL-8BDescriptor=Medical / Treatment Information2026.05 | 36.94 | 92.55 | |
| QwenSafeDescriptor=Simulated Gambling2026.05 | 35.14 | 97.44 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Realistic Violence2026.05 | 34.25 | 94.98 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Cartoon or Fantasy Violence2026.05 | 33.96 | 88.67 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Sexual Content or Nudity2026.05 | 33.33 | 90.16 | |
| Gemini-2.5-FlashDescriptor=Simulated Gambling2026.05 | 32.43 | 96.41 | |
| Qwen3-VL-8BDescriptor=Cartoon or Fantasy Violence2026.05 | 28.3 | 87.82 | |
| Qwen3-VL-8BDescriptor=Sexual Content or Nudity2026.05 | 27.54 | 89.4 | |
| Qwen3-VL-8BDescriptor=Horror or Fear Themes2026.05 | 26.09 | 91.06 | |
| Qwen3-VL-8BDescriptor=Realistic Violence2026.05 | 26.03 | 94.39 | |
| QwenSafeDescriptor=Real Gambling2026.05 | 24.56 | 95.46 | |
| Qwen3-VL-8BDescriptor=Mature or Suggestive Themes2026.05 | 24.38 | 76.54 | |
| Qwen3-VL-8BDescriptor=Average2026.05 | 24.1 | 89.99 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Profanity or Crude Humour2026.05 | 22.87 | 84.54 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Alcohol, Tobacco or Drug Use or References2026.05 | 16.87 | 85.52 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Mature or Suggestive Themes2026.05 | 16.25 | 74.65 | |
| Qwen3-VL-8BDescriptor=Simulated Gambling2026.05 | 16.22 | 96.82 | |
| Qwen3-VL-8BDescriptor=Unrestricted Web Access2026.05 | 16.19 | 90.87 | |
| LLaVA-v1.6-mistral-7b-hfDescriptor=Real Gambling2026.05 | 15.79 | 95.06 | |
| Gemini-2.5-FlashDescriptor=Real Gambling2026.05 | 15.79 | 93.13 | |
| Qwen3-VL-8BDescriptor=Alcohol, Tobacco or Drug Use or References2026.05 | 13.86 | 85.07 | |
| Qwen3-VL-8BDescriptor=Real Gambling2026.05 | 12.28 | 94.87 | |
| Qwen3-VL-8BDescriptor=Profanity or Crude Humour2026.05 | 10.11 | 82.43 |