Video Retrieval on Youcook2
36.4R@1Qwen3-VL-emb-8B + DS2M
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-VL-emb-8B + DS2MModel Type=Embedding Model, Fine-tuning Data=DS2M2026.04 | 36.4 | 58 | |
| Qwen3-VL-8B + DS2MModel Type=Instruct Model, Fine-tuning Data=DS2M2026.04 | 35.4 | 63 | |
| Qwen3-VL-8B + Ego4D + DS2MModel Type=Instruct Model, Fine-tuning Data=Ego4D + DS2M2026.04 | 34.7 | 63.1 | |
| Qwen3-VL-emb-8BModel Type=Embedding Model2026.04 | 31 | 58.7 | |
| Qwen3-VL-8B + Ego4DModel Type=Instruct Model, Fine-tuning Data=Ego4D2026.04 | 15.9 | 36 | |
| GRU+SSAPre-training=HowTo100M subset2021.05 | 10.9 | 28.4 | |
| VLM2Vec-V2.0Model Type=Comparison Model2026.04 | 10.6 | — | |
| GME-7BModel Type=Comparison Model2026.04 | 9.1 | — | |
| GRU+ResNeXt-101Pre-training=HowTo100M subset2021.05 | 8.2 | 24.5 | |
| LamRA-Qwen2.5Model Type=Comparison Model2026.04 | 7.5 | — | |
| Qwen3-VL-8BModel Type=Instruct Model2026.04 | 6.4 | 18.3 | |
| GRU+SSA2021.05 | 5.5 | 15.9 | |
| HGLMM FV CCA2021.05 | 4.6 | 14.3 | |
| GRU+ResNeXt-1012021.05 | 4.2 | 13.7 | |
| Random2021.05 | 0.03 | 0.15 |