Video Temporal Grounding on Charades-STA g43_s7 (test)
60.6R@0.5Qwen2-VL-7B + T2SGrid-FT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Qwen2-VL-7B + T2SGrid-FTModel Category=General Vision-LLMs, Framework=T2SGrid-FT2026.03 | 60.6 | 76.9 | 35.9 | 53.2 | |
| TimeSuiteModel Category=VTG-Tuned Vision-LLMs2026.03 | 48.7 | 69.9 | 24 | — | |
| Qwen3-VL-8B + T2SGridModel Category=General Vision-LLMs, Framework=T2SGrid2026.03 | 47 | 71.4 | 20.7 | 44.9 | |
| Qwen2-VL-7B + T2SGridModel Category=General Vision-LLMs, Framework=T2SGrid2026.03 | 46.7 | 70.1 | 20.1 | 44.3 | |
| Qwen3-VL-8BModel Category=General Vision-LLMs2026.03 | 43.4 | 69.3 | 17.5 | 43.1 | |
| NumProModel Category=VTG-Tuned Vision-LLMs2026.03 | 42 | 63.8 | 20.6 | 41.4 | |
| TRACEModel Category=VTG-Tuned Vision-LLMs2026.03 | 40.3 | — | 19.4 | — | |
| GPT-4o + T2SGridModel Category=General Vision-LLMs, Framework=T2SGrid2026.03 | 36.7 | 57.3 | 14.8 | 36.9 | |
| VTG-LLMModel Category=VTG-Tuned Vision-LLMs2026.03 | 33.8 | 52 | 15.7 | — | |
| GPT-4oModel Category=General Vision-LLMs2026.03 | 32 | 55 | 11.5 | 35.4 | |
| HawkEyeModel Category=VTG-Tuned Vision-LLMs2026.03 | 31.4 | 50.6 | 14.5 | 33.7 | |
| GroundingGPTModel Category=VTG-Tuned Vision-LLMs2026.03 | 29.6 | — | 11.9 | — | |
| VTimeLLMModel Category=VTG-Tuned Vision-LLMs2026.03 | 27.5 | 51 | 11.4 | 31.2 | |
| LLaVA-OneVision-1.5-8B + T2SGridModel Category=General Vision-LLMs, Framework=T2SGrid2026.03 | 26.3 | 45 | 11.9 | 28.8 | |
| MomentorModel Category=VTG-Tuned Vision-LLMs2026.03 | 23 | 42.9 | 12.4 | 29.3 | |
| TimeChatModel Category=VTG-Tuned Vision-LLMs2026.03 | 22.9 | 47.7 | 12.5 | 30.6 | |
| LLaVA-OneVision-1.5-8BModel Category=General Vision-LLMs2026.03 | 6.7 | 19.8 | 2.3 | 14.5 | |
| Qwen2-VL-7BModel Category=General Vision-LLMs2026.03 | 5.4 | 8.7 | 2.4 | 7.9 |