Instructional Temporal Answer Grounding in Video on CMIVQA (test)
63.33R@0.3GPT-3.5
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-3.5Family=LLM, Training Setting=zero-shot2026.06 | 63.33 | 47 | 25.67 | 43.87 | |
| CACRFamily=LVLM, Training Setting=supervised2026.06 | 62.5 | 48.52 | 33.82 | 45.37 | |
| GPT-3.5 (CoT)Family=LLM-CoT, Training Setting=zero-shot2026.06 | 56.33 | 40 | 20.33 | 39.16 | |
| Ouc AISFamily=VLP, Training Setting=supervised2026.06 | 50.88 | 35.42 | 20.54 | 36.37 | |
| SETAGFamily=VLP, Training Setting=supervised2026.06 | 47.75 | 32.09 | 18.98 | 33.89 | |
| MutualSLFamily=VLP, Training Setting=supervised2026.06 | 46.89 | 30.92 | 17.91 | 33.42 | |
| TimescopeFamily=LVLM, Training Setting=supervised2026.06 | 43.75 | 23.39 | 8.06 | 31.61 | |
| Time-R1Family=LVLM, Training Setting=supervised2026.06 | 43.45 | 28.36 | 15.09 | 32.06 | |
| TimeZeroFamily=LVLM, Training Setting=zero-shot2026.06 | 42.58 | 25.66 | 11.86 | 29.22 | |
| Ask2LocFamily=LVLM, Training Setting=supervised2026.06 | 42.46 | 25.04 | 17.14 | 31.37 | |
| VTPSLFamily=VLP, Training Setting=supervised2026.06 | 40.55 | 29.11 | 14.54 | 28.98 | |
| VTG-R1Family=LVLM, Training Setting=supervised2026.06 | 38.95 | 20.35 | 7.22 | 25.93 | |
| TFVTGFamily=LVLM, Training Setting=zero-shot2026.06 | 29.94 | 16.01 | 9.88 | 22.44 | |
| Qwen2.5-VL-7B-InstructFamily=LVLM, Training Setting=zero-shot2026.06 | 15.38 | 9.83 | 4.7 | 12.21 | |
| Random Mode2026.06 | 5.71 | 4.65 | 3.58 | 3.97 |