Average Context Length on NExT-GQA for Video Question Answering
55,136.3Average Context Length (tokens)Video-o3
Evaluation Results
| Method | Links | |
|---|---|---|
| Video-o3Max retrieval / injection=82026.05 | 55,136.3 | |
| LongVT-RFTMax retrieval / injection=52026.05 | 16,389 | |
| VideoZoomerMax retrieval / injection=42026.05 | 12,254.9 | |
| VideoChat-R1.5-7BMax retrieval / injection=22026.05 | 8,897.5 | |
| LOVE-R1-stage3Max retrieval / injection=22026.05 | 8,480.5 | |
| DynFrame-8BMax retrieval / injection=12026.05 | 6,734.5 | |
| Video-R1-7BMax retrieval / injection=02026.05 | 3,893.2 | |
| Qwen2.5-VL-7BMax retrieval / injection=02026.05 | 2,910.9 |