Closed-ended Task T-8 on DanmakuTPP-QA
47.17ACCQwen2.5-VL-72B
Evaluation results
| Method | Links | |
|---|---|---|
| Qwen2.5-VL-72BModel Category=MLLMs2026.02 | 47.17 | |
| MM-TPPModel Category=Finetuned2026.02 | 44 | |
| Qwen2.5-72B-InstructModel Category=LLMs2026.02 | 43.83 | |
| Qwen3-30B-A3BModel Category=LLMs2026.02 | 43.67 | |
| Qwen2.5-VL-3BModel Category=Finetuned2026.02 | 43 | |
| Qwen3-8BModel Category=LLMs2026.02 | 41.5 | |
| Qwen2.5-32B-InstructModel Category=LLMs2026.02 | 38.17 | |
| Gemma3-27BModel Category=MLLMs2026.02 | 36.17 | |
| DeepSeek-V3Model Category=LLMs2026.02 | 34.5 | |
| Llama-3.3-70B-InstructModel Category=LLMs2026.02 | 33.33 | |
| Qwen2.5-7B-InstructModel Category=LLMs2026.02 | 32.67 | |
| Qwen3-235B-A22BModel Category=LLMs2026.02 | 32.5 | |
| Qwen2.5-VL-7BModel Category=MLLMs2026.02 | 22.17 | |
| Qwen2.5-VL-32BModel Category=MLLMs2026.02 | 22.17 |