Spatio-temporal reasoning on STCR
100AccuracyHuman Perception
Evaluation Results
| Method | Links | |
|---|---|---|
| Human PerceptionWinSize=102026.04 | 100 | |
| Human PerceptionWinSize=112026.04 | 100 | |
| Human PerceptionWinSize=≥122026.04 | 100 | |
| Gemini-2.5-ProWinSize=≥122026.04 | 100 | |
| Human PerceptionWinSize=82026.04 | 99.1 | |
| Human PerceptionWinSize=92026.04 | 99.1 | |
| Seed1.6WinSize=≥122026.04 | 98.3 | |
| Human PerceptionWinSize=72026.04 | 98.2 | |
| Human PerceptionWinSize=52026.04 | 97.3 | |
| Human PerceptionWinSize=62026.04 | 97.3 | |
| Doubao-1.5-proWinSize=≥122026.04 | 96 | |
| Doubao-1.5-proWinSize=112026.04 | 90 | |
| Gemini-2.5-ProWinSize=112026.04 | 89.7 | |
| STCR-TagWinSize=≥122026.04 | 88.7 | |
| STCR-TagWinSize=102026.04 | 88.6 | |
| STCR-TagWinSize=112026.04 | 88.6 | |
| STCR-TagWinSize=92026.04 | 87.7 | |
| STCR-TagWinSize=82026.04 | 87.1 | |
| Seed1.6WinSize=112026.04 | 86.8 | |
| STCR-TagWinSize=72026.04 | 86.2 | |
| Doubao-1.5-proWinSize=102026.04 | 86 | |
| STCR-TagWinSize=62026.04 | 84.8 | |
| Seed1.6WinSize=92026.04 | 83.2 | |
| Seed1.6WinSize=102026.04 | 82.9 | |
| STCR-TagWinSize=52026.04 | 82.4 | |
| DeepSeek-VL2WinSize=≥122026.04 | 81 | |
| Gemini-2.5-ProWinSize=102026.04 | 80.9 | |
| GPT-4oWinSize=92026.04 | 79.2 | |
| GPT-4oWinSize=102026.04 | 79.2 | |
| GPT-4oWinSize=112026.04 | 79.2 | |
| GPT-4oWinSize=≥122026.04 | 79.2 | |
| GPT-o4mini-highWinSize=≥122026.04 | 78.6 | |
| GPT-o4mini-highWinSize=112026.04 | 78.1 | |
| GPT-4oWinSize=82026.04 | 77.8 | |
| STCR-CoTWinSize=≥122026.04 | 77.8 | |
| DeepSeek-VL2WinSize=112026.04 | 77.4 | |
| GPT-o4mini-highWinSize=102026.04 | 77.3 | |
| Gemini-2.5-ProWinSize=92026.04 | 77.2 | |
| GPT-4oWinSize=72026.04 | 76.7 | |
| STCR-CoTWinSize=112026.04 | 76.7 | |
| Doubao-1.5-proWinSize=62026.04 | 76 | |
| STCR-CoTWinSize=102026.04 | 75.6 | |
| GPT-o4mini-highWinSize=92026.04 | 75.5 | |
| Gemini-2.5-ProWinSize=62026.04 | 75 | |
| Gemini-2.5-ProWinSize=82026.04 | 75 | |
| STCR-CoTWinSize=92026.04 | 74.5 | |
| STCR-CoTWinSize=82026.04 | 73.7 | |
| Seed1.6WinSize=82026.04 | 73.3 | |
| GPT-o4miniWinSize=72026.04 | 72.8 | |
| Gemini-2.5-ProWinSize=72026.04 | 72.8 | |
| GPT-o4mini-highWinSize=62026.04 | 72.7 | |
| GPT-o4mini-highWinSize=82026.04 | 72.3 | |
| STCR-CoTWinSize=72026.04 | 72.2 | |
| GPT-o4mini-highWinSize=72026.04 | 71.8 | |
| GPT-4oWinSize=62026.04 | 71.7 | |
| Gemini-2.5-ProWinSize=52026.04 | 71.3 | |
| GPT-4oWinSize=52026.04 | 71.1 | |
| Seed1.6WinSize=72026.04 | 70.6 | |
| GPT-o4mini-highWinSize=52026.04 | 70.5 | |
| GPT-o4miniWinSize=62026.04 | 70.3 | |
| GPT-o4miniWinSize=≥122026.04 | 70.3 | |
| STCR-CoTWinSize=62026.04 | 70.2 | |
| GPT-o4miniWinSize=52026.04 | 70 | |
| DeepSeek-VL2WinSize=102026.04 | 69.1 | |
| STCR-CoTWinSize=52026.04 | 68.3 | |
| DeepSeek-VL2WinSize=82026.04 | 67.8 | |
| GPT-o4miniWinSize=82026.04 | 67.7 | |
| GPT-o4miniWinSize=112026.04 | 67.7 | |
| GPT-o4miniWinSize=102026.04 | 67.5 | |
| GPT-o3WinSize=52026.04 | 67.2 | |
| DeepSeek-VL2WinSize=92026.04 | 67 | |
| DeepSeek-VL2WinSize=62026.04 | 66.9 | |
| Seed1.6WinSize=62026.04 | 66.8 | |
| GPT-o3WinSize=≥122026.04 | 66.6 | |
| Seed1.6WinSize=52026.04 | 66.2 | |
| DeepSeek-VL2WinSize=72026.04 | 65.2 | |
| DeepSeek-VL2WinSize=52026.04 | 64.4 | |
| GPT-o3WinSize=102026.04 | 64 | |
| Doubao-1.5-proWinSize=52026.04 | 64 | |
| GPT-o3WinSize=112026.04 | 62 | |
| GPT-o3WinSize=82026.04 | 61.6 | |
| GPT-o3WinSize=62026.04 | 61.4 | |
| GPT-o4miniWinSize=92026.04 | 60 | |
| GPT-o3WinSize=72026.04 | 59.6 | |
| GPT-o3WinSize=92026.04 | 59.6 | |
| Qwen2.5-VL-7BWinSize=82026.04 | 57.3 | |
| Doubao-1.5-proWinSize=92026.04 | 56.9 | |
| Qwen2.5-VL-7BWinSize=≥122026.04 | 56.5 | |
| NVILA-8BWinSize=82026.04 | 56.1 | |
| Qwen2.5-VL-7BWinSize=102026.04 | 56.1 | |
| Doubao-1.5-proWinSize=72026.04 | 56 | |
| Qwen2.5-VL-7BWinSize=112026.04 | 56 | |
| Qwen2.5-VL-7BWinSize=62026.04 | 55.4 | |
| LLaVA-OneVision-7BWinSize=62026.04 | 55.4 | |
| Qwen2.5-VL-7BWinSize=92026.04 | 55.2 | |
| NVILA-8BWinSize=≥122026.04 | 55 | |
| Qwen2.5-VL-7BWinSize=72026.04 | 55 | |
| LLaVA-OneVision-7BWinSize=52026.04 | 54.8 | |
| LLaVA-OneVision-7BWinSize=72026.04 | 53.5 | |
| Qwen2.5-VL-7BWinSize=52026.04 | 53.1 |