Action Recognition on SSv2 Small
76.9AccuracyFSAR-Qwen3
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| FSAR-Qwen3Pre-training=Qwen3-VL-8B, Knowledge Prompt=Known, Shot=5-shot2026.03 | 76.9 | — | — | — | |
| FSAR-Qwen3Pre-training=Qwen3-VL-2B, Knowledge Prompt=Known, Shot=5-shot2026.03 | 76.3 | — | — | — | |
| FSAR-LLaVAPre-training=Video-LLaVA-7B, Knowledge Prompt=Known, Shot=5-shot2026.03 | 76.1 | — | — | — | |
| FSAR-Qwen2Pre-training=Qwen2-VL-7B, Knowledge Prompt=Known, Shot=5-shot2026.03 | 75.9 | — | — | — | |
| FSAR-Qwen3Pre-training=Qwen3-VL-8B, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 71.9 | — | — | — | |
| FSAR-Qwen3Pre-training=Qwen3-VL-2B, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 71.7 | — | — | — | |
| FSAR-Qwen2Pre-training=Qwen2-VL-7B, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 71.3 | — | — | — | |
| FSAR-LLaVAPre-training=Video-LLaVA-7B, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 71.1 | — | — | — | |
| Task-AdapterPre-training=CLIP-ViT-B, Shot=5-shot2026.03 | 70.2 | — | — | — | |
| FSAR-Qwen3Pre-training=Qwen3-VL-8B, Knowledge Prompt=Known, Shot=1-shot2026.03 | 68.4 | — | — | — | |
| FSAR-Qwen3Pre-training=Qwen3-VL-2B, Knowledge Prompt=Known, Shot=1-shot2026.03 | 67.1 | — | — | — | |
| FSAR-Qwen2Pre-training=Qwen2-VL-7B, Knowledge Prompt=Known, Shot=1-shot2026.03 | 66.8 | — | — | — | |
| EMP-NetPre-training=CLIP-ViT-B, Shot=5-shot2026.03 | 65.7 | — | — | — | |
| FSAR-LLaVAPre-training=Video-LLaVA-7B, Knowledge Prompt=Known, Shot=1-shot2026.03 | 65.5 | — | — | — | |
| CLIP-CPM2CPre-training=CLIP-ViT-B, Shot=5-shot2026.03 | 62.6 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-8B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=5-shot2026.03 | 62.2 | — | — | — | |
| MVP-shotPre-training=CLIP-ViT-B, Shot=5-shot2026.03 | 62 | — | — | — | |
| Huang et al.Pre-training=INet-RN50, Shot=5-shot2026.03 | 61.8 | — | — | — | |
| CLIP-FSARPre-training=CLIP-ViT-B, Shot=5-shot2026.03 | 61.8 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-8B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 61.5 | — | — | — | |
| CapFSARPre-training=BLIP-ViT-B, Shot=5-shot2026.03 | 61.1 | — | — | — | |
| DISTFine-tuning=PEFT, Setting=5-way 1-shot2026.02 | 60.7 | — | — | — | |
| Task-AdapterFine-tuning=PEFT, Setting=5-way 1-shot2026.02 | 60.2 | — | — | — | |
| Task-AdapterPre-training=CLIP-ViT-B, Shot=1-shot2026.03 | 60.2 | — | — | — | |
| MA-FSARFine-tuning=PEFT, Setting=5-way 1-shot2026.02 | 59.1 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-2B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=5-shot2026.03 | 57.8 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-2B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 57.2 | — | — | — | |
| EMP-NetPre-training=CLIP-ViT-B, Shot=1-shot2026.03 | 57.1 | — | — | — | |
| TRXPre-training=INet-RN50, Shot=5-shot2026.03 | 56.7 | — | — | — | |
| Qwen2-VLPre-training=Qwen2-VL-7B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=5-shot2026.03 | 56.3 | — | — | — | |
| MoLo (OTAM)Pre-training=INet-RN50, Shot=5-shot2026.03 | 56.2 | — | — | — | |
| HyRSMPre-training=INet-RN50, Shot=5-shot2026.03 | 56.1 | — | — | — | |
| HCLPre-training=INet-RN50, Shot=5-shot2026.03 | 55.4 | — | — | — | |
| MVP-shotPre-training=CLIP-ViT-B, Shot=1-shot2026.03 | 55.4 | — | — | — | |
| STRMPre-training=INet-RN50, Shot=5-shot2026.03 | 55.3 | — | — | — | |
| Qwen2-VLPre-training=Qwen2-VL-7B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 55.1 | — | — | — | |
| D2ST-AdapterFine-tuning=PEFT, Setting=5-way 1-shot2026.02 | 55 | — | — | — | |
| FSAR-Qwen3Pre-training=Qwen3-VL-8B, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 55 | — | — | — | |
| CLIP-FSARPre-training=CLIP-ViT-B, Shot=1-shot2026.03 | 54.6 | — | — | — | |
| FSAR-Qwen3Pre-training=Qwen3-VL-2B, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 54.2 | — | — | — | |
| FSAR-Qwen2Pre-training=Qwen2-VL-7B, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 53.6 | — | — | — | |
| HR2G-shotBackbone=ResNet-based CLIP, Few-shot Setting=5-way 1-shot2025.04 | 53.1 | — | — | — | |
| CLIP-CPM2CPre-training=CLIP-ViT-B, Shot=1-shot2026.03 | 52.3 | — | — | — | |
| CLIP-FSARBackbone=ResNet-based CLIP, Few-shot Setting=5-way 1-shot2025.04 | 52 | — | — | — | |
| MVP-shotBackbone=ResNet-based CLIP, Few-shot Setting=5-way 1-shot2025.04 | 51.2 | — | — | — | |
| FSAR-LLaVAPre-training=Video-LLaVA-7B, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 51 | — | — | — | |
| Video-LLaVAPre-training=Video-LLaVA-7B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=5-shot2026.03 | 48.5 | — | — | — | |
| OTAMPre-training=INet-RN50, Shot=5-shot2026.03 | 48 | — | — | — | |
| Video-LLaVAPre-training=Video-LLaVA-7B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=5-shot2026.03 | 47.6 | — | — | — | |
| CapFSARPre-training=BLIP-ViT-B, Shot=1-shot2026.03 | 45.8 | — | — | — | |
| Huang et al.Pre-training=INet-RN50, Shot=1-shot2026.03 | 42.6 | — | — | — | |
| MoLo (OTAM)Pre-training=INet-RN50, Shot=1-shot2026.03 | 41.9 | — | — | — | |
| HyRSMPre-training=INet-RN50, Shot=1-shot2026.03 | 40.6 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-8B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=1-shot2026.03 | 39.8 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-8B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 39 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-2B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=1-shot2026.03 | 38.8 | — | — | — | |
| HCLPre-training=INet-RN50, Shot=1-shot2026.03 | 38.7 | — | — | — | |
| Qwen2-VLPre-training=Qwen2-VL-7B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=1-shot2026.03 | 38.7 | — | — | — | |
| Qwen3-VLPre-training=Qwen3-VL-2B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 37.2 | — | — | — | |
| STRMPre-training=INet-RN50, Shot=1-shot2026.03 | 37.1 | — | — | — | |
| OTAMPre-training=INet-RN50, Shot=1-shot2026.03 | 36.4 | — | — | — | |
| Qwen2-VLPre-training=Qwen2-VL-7B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 36.1 | — | — | — | |
| TRXPre-training=INet-RN50, Shot=1-shot2026.03 | 36 | — | — | — | |
| Video-LLaVAPre-training=Video-LLaVA-7B, Mode=Frozen Avg, Knowledge Prompt=Known, Shot=1-shot2026.03 | 33.7 | — | — | — | |
| Video-LLaVAPre-training=Video-LLaVA-7B, Mode=Frozen Avg, Knowledge Prompt=Unknown, Shot=1-shot2026.03 | 32.7 | — | — | — | |
| A2M2-NetPre-training=INet-RN50, K-way=5-way2025.04 | — | 42.9 | — | 60.5 | |
| CapFSARPre-training=CLIP-ViT-B, K-way=5-way2025.04 | — | 45.9 | — | 59.9 | |
| CLIP-CPM2CPre-training=CLIP-RN502025.12 | — | 51.5 | 54.8 | 57.1 | |
| CLIP-CPM2CPre-training=CLIP-ViT-B2025.12 | — | 52.3 | 59.9 | 62.6 | |
| CLIP-FreezePre-training=CLIP-ViT-B, K-way=5-way2025.04 | — | 29.5 | — | 42.5 | |
| CLIP-FSARPre-training=CLIP-RN502025.12 | — | 52 | 54 | 55.8 | |
| CLIP-FSARPre-training=CLIP-ViT-B2025.12 | — | 54.5 | 58.6 | 61.8 | |
| CLIP-FSARBackbone=CLIP-ViT-B/162025.12 | — | 54.6 | — | 61.8 | |
| CLIP-FSARPre-training=CLIP-RN50, K-way=5-way2025.04 | — | 52 | — | 55.8 | |
| CLIP-FSARPre-training=CLIP-ViT-B, K-way=5-way2025.04 | — | 54.6 | — | 61.8 | |
| CLIP-M2DFPre-training=CLIP-RN502025.12 | — | 51.6 | 56.3 | 60.4 | |
| CLIP-M2DFPre-training=CLIP-ViT-B2025.12 | — | 52 | 58.9 | 62.8 | |
| CLIP-SPMPre-training=CLIP-RN502025.12 | — | 50.8 | 54.9 | 58.5 | |
| CLIP-SPMPre-training=CLIP-ViT-B2025.12 | — | 57.8 | 62.4 | 66.2 | |
| CMNPre-training=INet-RN50, K-way=5-way2025.04 | — | 34.4 | — | 43.8 | |
| D2ST-AdapterBackbone=CLIP-ViT-B/162025.12 | — | 55 | — | 69.3 | |
| EMP-NetBackbone=CLIP-ViT-B/162025.12 | — | 57.1 | — | 65.7 | |
| EMP-NetPre-training=CLIP-ViT-B, K-way=5-way2025.04 | — | 57.1 | — | 65.7 | |
| HCLBackbone=IN-RN502025.12 | — | 38.7 | — | 55.4 | |
| HCLPre-training=INet-RN50, K-way=5-way2025.04 | — | 38.9 | — | 55.4 | |
| HR2G-shotPre-training=CLIP-ViT-B, K-way=5-way2025.04 | — | 60.2 | — | 66 | |
| HyRSMPre-training=INet-RN502025.12 | — | 40.6 | 52.3 | 56.1 | |
| HyRSMBackbone=IN-RN502025.12 | — | 40.6 | — | 56.1 | |
| HyRSMPre-training=INet-RN50, K-way=5-way2025.04 | — | 40.6 | — | 56.1 | |
| MA-FSARBackbone=CLIP-ViT-B/162025.12 | — | 59.1 | — | 64.5 | |
| MA-FSARPre-training=CLIP-ViT-B, K-way=5-way2025.04 | — | 59.1 | — | 64.5 | |
| MoLoPre-training=INet-RN502025.12 | — | 42.7 | 52.9 | 56.4 | |
| MoLoPre-training=INet-RN50, K-way=5-way2025.04 | — | 41.9 | — | 56.2 | |
| MVP-shotBackbone=CLIP-ViT-B/162025.12 | — | 55.4 | — | 62 | |
| MVP-shotPre-training=CLIP-ViT-B, K-way=5-way2025.04 | — | 55.4 | — | 62 | |
| OTAMPre-training=INet-RN502025.12 | — | 38.9 | — | 48.1 | |
| OTAMBackbone=IN-RN502025.12 | — | 36.4 | — | 48 | |
| STENPre-training=CLIP-RN502025.12 | — | 36.4 | — | 57.3 | |
| STENPre-training=CLIP-ViT-B2025.12 | — | 38.9 | — | 58.2 | |
| STRMBackbone=IN-RN502025.12 | — | 37.1 | — | 55.3 |