# Few-shot Video Recognition on Something-Something V2 (K=2, 4, 8, 16) Source: [Few-shot Video Recognition on Something-Something V2 (K=2, 4, 8, 16)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:06:53.788Z Evaluation of few-shot video recognition performance on the Something-Something V2 dataset, specifically reporting Top-1 Accuracy across 2-shot, 4-shot, 8-shot, and 16-shot settings. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Few-shot video recognition | | Dataset (including any stated split) | Something-Something V2 | | Metric | Top-1 Acc (K=2) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 13 of 13 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Top-1 Acc (K=2)](): Higher is better; unit: PERCENT. - [Top-1 Acc (K=4)](): Higher is better; unit: PERCENT. - [Top-1 Acc (K=8)](): Higher is better; unit: PERCENT. - [Top-1 Acc (K=16)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Top-1 Acc (K=2) | Result date | Report paper | | --- | --- | --- | --- | | Zero-ViCLIP + MSTA + Lcc | 9.1 | Not recorded | 68fa77771e209131cea73519 | | Zero-ViCLIP + MSTA | 9 | Not recorded | 68fa77771e209131cea73519 | | ViCLIP | 8.7 | Not recorded | 68fa77771e209131cea73519 | | Zero-ViCLIP + AdaptFormer | 8.5 | Not recorded | 68fa77771e209131cea73519 | | Zero-ViCLIP + LORA | 8.4 | Not recorded | 68fa77771e209131cea73519 | | MOTE | 7.4 | Not recorded | 68fa77771e209131cea73519 | | OST | 7.3 | Not recorded | 68fa77771e209131cea73519 | | Vanilla CLIP | 6.7 | Not recorded | 68fa77771e209131cea73519 | | ViFi-CLIP | 6.5 | Not recorded | 68fa77771e209131cea73519 | | Zero-ViCLIP | 5.1 | Not recorded | 68fa77771e209131cea73519 | | A5 | 4.5 | Not recorded | 68fa77771e209131cea73519 | | XCLIP | 4.4 | Not recorded | 68fa77771e209131cea73519 | | Vanilla CLIP | 2.8 | Not recorded | 68fa77771e209131cea73519 | ### Zero-ViCLIP + MSTA + Lcc Top-1 Acc (K=2): 9.1. Result ID: 699493cb3a6c3148491e2141. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"GFLOPs":"179.84","Transfer learning protocol":"Efficient transfer pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### Zero-ViCLIP + MSTA Top-1 Acc (K=2): 9. Result ID: 699493cb3a6c3148491e2140. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"GFLOPs":"179.84","Transfer learning protocol":"Efficient transfer pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### ViCLIP Top-1 Acc (K=2): 8.7. Result ID: 699493cb3a6c3148491e213c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"GFLOPs":"161","Transfer learning protocol":"Tuning pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### Zero-ViCLIP + AdaptFormer Top-1 Acc (K=2): 8.5. Result ID: 699493cb3a6c3148491e213e. [Reporting paper]() Recorded evaluation configuration: {"GFLOPs":"179.80","Transfer learning protocol":"Efficient transfer pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### Zero-ViCLIP + LORA Top-1 Acc (K=2): 8.4. Result ID: 699493cb3a6c3148491e213f. [Reporting paper]() Recorded evaluation configuration: {"GFLOPs":"179.80","Transfer learning protocol":"Efficient transfer pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### MOTE Top-1 Acc (K=2): 7.4. Result ID: 699493cb3a6c3148491e213b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"GFLOPs":"141","Transfer learning protocol":"Tuning pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### OST Top-1 Acc (K=2): 7.3. Result ID: 699493cb3a6c3148491e213a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Transfer learning protocol":"Tuning pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### Vanilla CLIP Top-1 Acc (K=2): 6.7. Result ID: 699493cb3a6c3148491e2138. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"GFLOPs":"282","Transfer learning protocol":"Tuning pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### ViFi-CLIP Top-1 Acc (K=2): 6.5. Result ID: 699493cb3a6c3148491e2139. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"GFLOPs":"281","Transfer learning protocol":"Tuning pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### Zero-ViCLIP Top-1 Acc (K=2): 5.1. Result ID: 699493cb3a6c3148491e213d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"GFLOPs":"161","Transfer learning protocol":"Efficient transfer pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### A5 Top-1 Acc (K=2): 4.5. Result ID: 699493cb3a6c3148491e2137. [Reporting paper]() Recorded evaluation configuration: {"GFLOPs":"284","Transfer learning protocol":"Adapting pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### XCLIP Top-1 Acc (K=2): 4.4. Result ID: 699493cb3a6c3148491e2136. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"GFLOPs":"145","Transfer learning protocol":"Adapting pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. ### Vanilla CLIP Top-1 Acc (K=2): 2.8. Result ID: 699493cb3a6c3148491e2135. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"GFLOPs":"281","Transfer learning protocol":"Zero-shot pre-trained models","Multi-view inference":"true"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.