Procedural Temporal Understanding on MuirBench (test)
65.04Overall ScoreTPRU-7B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TPRU-7BBackbone=7B2026.02 | 65.04 | 34.38 | |
| Qwen2.5-Omni-7BBackbone=Qwen2.5-7B2026.02 | 59.11 | 18.75 | |
| Long-VITA-16KContext Window=16K2026.02 | 53.07 | 17.19 | |
| LLaVA-Video-Qwen2-72BBackbone=Qwen2-72B2026.02 | 41.81 | 10.94 | |
| LLaVA-Video-Qwen2-7BBackbone=Qwen2-7B2026.02 | 37.88 | 15.63 | |
| SmolVLM2-256M-VideoParameters=256M2026.02 | 27.92 | 21.88 | |
| SmolVLM2-500M-VideoParameters=500M2026.02 | 25.92 | 7.8 |