Information Analysis and Processing on MARS Overall (Evaluation Set)
1.93Average Ranking (AI GPT-5)Ours
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| OursParam.=1800B2025.11 | 1.93 | 1.33 | |
| Qwen3-235B-AParam.=22B2025.11 | 2.77 | 3.77 | |
| Llama-VisionParam.=11B2025.11 | 5.23 | 6.3 | |
| Llama-3.2-11BParam.=11B2025.11 | 5.77 | 6.2 | |
| Llama-4-ScoutParam.=17B2025.11 | 5.77 | 6.4 | |
| Qwen-VLParam.=7B2025.11 | 6.18 | 7.6 | |
| Llama-3.2-90BParam.=90B2025.11 | 6.77 | 5.9 | |
| GPT-ossParam.=20B2025.11 | 6.8 | 6.37 | |
| LLaVA-1.5Param.=7B2025.11 | 8.03 | 6.33 | |
| Qwen2-VLParam.=72B2025.11 | 8.12 | 7.2 | |
| Qwen2.5-VLParam.=72B2025.11 | 8.43 | 8.33 |