# Video Action Recognition on SS v2 (ACC@1)

Source: [Video Action Recognition on SS v2 (ACC@1)](<https://www.sota2.com/research/sota/video-action-recognition-on-ss-v2-acc-at-1>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-13T07:57:43.513Z

Evaluation of Video Action Recognition performance on the Something-Something v2 (SS v2) dataset, specifically reporting Top-1 Accuracy (ACC@1).

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 | Video Action Recognition |
| Dataset (including any stated split) | SS v2 |
| Metric | Top-1 Accuracy (SS v2) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 26 of 26 |
| 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 Accuracy (SS v2)](<https://www.sota2.com/research/sota/video-action-recognition-on-ss-v2-acc-at-1.md?metric=Top-1%20Accuracy%20(SS%20v2)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Top-1 Accuracy (SS v2) | Result date | Report paper |
| --- | --- | --- | --- |
| VideoMAE V2 | 76.8 | Not recorded | 68fa7b991e209131cea89646 |
| V-JEPA2-L | 73.7 | Not recorded | 69b39a370636e5bf4afbccff |
| V-JEPA-2 | 73.7 | Not recorded | 6a016288dbc836b87d82305a |
| LookWhen (70% sparse) | 72 | Not recorded | 6a016288dbc836b87d82305a |
| VideoMAEv2 | 71.2 | Not recorded | 6a016288dbc836b87d82305a |
| VideoMamba-M800e | 71 | Not recorded | 6a016288dbc836b87d82305a |
| UMT-B800e | 70.8 | Not recorded | 6a016288dbc836b87d82305a |
| VideoMAE | 70.8 | Not recorded | 6a016288dbc836b87d82305a |
| VideoMAE + RLT (τ=0.1) | 70.2 | Not recorded | 6a016288dbc836b87d82305a |
| InternVideo2 | 69.7 | Not recorded | 6a016288dbc836b87d82305a |
| VideoMAE + ToMe (r=64) | 69.7 | Not recorded | 6a016288dbc836b87d82305a |
| VideoMAE + vid-TLDR | 69.6 | Not recorded | 6a016288dbc836b87d82305a |
| VideoMambaPro | 69.4 | Not recorded | 6a016288dbc836b87d82305a |
| LookWhen (90% sparse) | 69.3 | Not recorded | 6a016288dbc836b87d82305a |
| OmniStream | 68.5 | Not recorded | 69b39a370636e5bf4afbccff |
| VideoMAE + LITE (K=0.3) | 68.3 | Not recorded | 6a016288dbc836b87d82305a |
| VideoSwin + STTS | 68.1 | Not recorded | 6a016288dbc836b87d82305a |
| NoFrame | 62.7 | Not recorded | 68fa7b991e209131cea89646 |
| TSM | 61.7 | Not recorded | 68fa7b991e209131cea89646 |
| DINOv3-L | 54 | Not recorded | 69b39a370636e5bf4afbccff |
| MoTIF-ST | 41.9 | Not recorded | 68fa7b991e209131cea89646 |
| MoTIF | 40.4 | Not recorded | 68fa7b991e209131cea89646 |
| MoTIF | 37.3 | Not recorded | 68fa7b991e209131cea89646 |
| Global CBM | 33.6 | Not recorded | 68fa7b991e209131cea89646 |
| MoTIF | 25.8 | Not recorded | 68fa7b991e209131cea89646 |
| Zero-shot | 2.2 | Not recorded | 68fa7b991e209131cea89646 |

### VideoMAE V2

Top-1 Accuracy (SS v2): 76.8. Result ID: 6a042eec47f34ce924e9e322.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f171e209131cea43db1.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### V-JEPA2-L

Top-1 Accuracy (SS v2): 73.7. Result ID: 69b3dba847f34ce9249e5ef6.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a370636e5bf4afbccff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d6a4.md>)

[Implementation](<https://github.com/facebookresearch/vjepa2>)

Recorded evaluation configuration: {"Model Size":"Large"}. This record does not assert independent reproduction.

### V-JEPA-2

Top-1 Accuracy (SS v2): 73.7. Result ID: 6a0184ab47f34ce924e2f851.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79651e209131cea7d6a4.md>)

[Implementation](<https://github.com/facebookresearch/vjepa2>)

Recorded evaluation configuration: {"Params (M)":"355","FLOPs (G×T×S)":"935×2×3"}. This record does not assert independent reproduction.

### LookWhen (70% sparse)

Top-1 Accuracy (SS v2): 72. Result ID: 6a0184ab47f34ce924e2f85e.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

Recorded evaluation configuration: {"Params (M)":"106","FLOPs (G×T×S)":"108×2×3","sparsity":"70%"}. This record does not assert independent reproduction.

### VideoMAEv2

Top-1 Accuracy (SS v2): 71.2. Result ID: 6a0184ab47f34ce924e2f855.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6f171e209131cea43db1.md>)

Recorded evaluation configuration: {"Params (M)":"87","FLOPs (G×T×S)":"180×2×3"}. This record does not assert independent reproduction.

### VideoMamba-M800e

Top-1 Accuracy (SS v2): 71. Result ID: 6a0184ab47f34ce924e2f856.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72831e209131cea56ea1.md>)

[Implementation](<https://github.com/OpenGVLab/VideoMamba>)

Recorded evaluation configuration: {"Params (M)":"74","FLOPs (G×T×S)":"202×2×3"}. This record does not assert independent reproduction.

### UMT-B800e

Top-1 Accuracy (SS v2): 70.8. Result ID: 6a0184ab47f34ce924e2f853.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72711e209131cea5652b.md>)

[Implementation](<https://github.com/OpenGVLab/unmasked_teacher>)

Recorded evaluation configuration: {"Params (M)":"87","FLOPs (G×T×S)":"180×2×3"}. This record does not assert independent reproduction.

### VideoMAE

Top-1 Accuracy (SS v2): 70.8. Result ID: 6a0184ab47f34ce924e2f854.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6da81e209131cea3b2fd.md>)

[Implementation](<https://github.com/MCG-NJU/VideoMAE>)

Recorded evaluation configuration: {"Params (M)":"87","FLOPs (G×T×S)":"180×2×3"}. This record does not assert independent reproduction.

### VideoMAE + RLT (τ=0.1)

Top-1 Accuracy (SS v2): 70.2. Result ID: 6a0184ab47f34ce924e2f85b.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75a21e209131cea691cb.md>)

Recorded evaluation configuration: {"Params (M)":"87","FLOPs (G×T×S)":"120×4×3"}. This record does not assert independent reproduction.

### InternVideo2

Top-1 Accuracy (SS v2): 69.7. Result ID: 6a0184ab47f34ce924e2f852.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74871e209131cea623f3.md>)

[Implementation](<https://github.com/OpenGVLab/InternVideo2>)

Recorded evaluation configuration: {"Params (M)":"1020","FLOPs (G×T×S)":"2500×2×3"}. This record does not assert independent reproduction.

### VideoMAE + ToMe (r=64)

Top-1 Accuracy (SS v2): 69.7. Result ID: 6a0184ab47f34ce924e2f85a.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6ea91e209131cea4130c.md>)

[Implementation](<http://github.com/facebookresearch/ToMe>)

Recorded evaluation configuration: {"Params (M)":"87","FLOPs (G×T×S)":"131×4×3"}. This record does not assert independent reproduction.

### VideoMAE + vid-TLDR

Top-1 Accuracy (SS v2): 69.6. Result ID: 6a0184ab47f34ce924e2f85c.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa72bc1e209131cea58413.md>)

[Implementation](<https://github.com/mlvlab/vid-TLDR>)

Recorded evaluation configuration: {"Params (M)":"87","FLOPs (G×T×S)":"57×unk"}. This record does not assert independent reproduction.

### VideoMambaPro

Top-1 Accuracy (SS v2): 69.4. Result ID: 6a0184ab47f34ce924e2f857.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75bb1e209131cea6998c.md>)

Recorded evaluation configuration: {"Params (M)":"72","FLOPs (G×T×S)":"183×4×3"}. This record does not assert independent reproduction.

### LookWhen (90% sparse)

Top-1 Accuracy (SS v2): 69.3. Result ID: 6a0184ab47f34ce924e2f85d.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

Recorded evaluation configuration: {"Params (M)":"106","FLOPs (G×T×S)":"40×2×3","sparsity":"90%"}. This record does not assert independent reproduction.

### OmniStream

Top-1 Accuracy (SS v2): 68.5. Result ID: 69b3dba847f34ce9249e5ef4.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a370636e5bf4afbccff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b39a370636e5bf4afbccff.md>)

[Implementation](<https://github.com/Go2Heart/OmniStream>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### VideoMAE + LITE (K=0.3)

Top-1 Accuracy (SS v2): 68.3. Result ID: 6a0184ab47f34ce924e2f859.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77981e209131cea743ea.md>)

Recorded evaluation configuration: {"Params (M)":"87","FLOPs (G×T×S)":"46×2×3"}. This record does not assert independent reproduction.

### VideoSwin + STTS

Top-1 Accuracy (SS v2): 68.1. Result ID: 6a0184ab47f34ce924e2f858.

[Reporting paper](<https://www.sota2.com/research/paper/6a016288dbc836b87d82305a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cfc1e209131cea371c7.md>)

Recorded evaluation configuration: {"Params (M)":"89","FLOPs (G×T×S)":"190×1×3"}. This record does not assert independent reproduction.

### NoFrame

Top-1 Accuracy (SS v2): 62.7. Result ID: 6a042eec47f34ce924e9e321.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69a11e209131cea23371.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### TSM

Top-1 Accuracy (SS v2): 61.7. Result ID: 6a042eec47f34ce924e9e320.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66381e209131cea0ee93.md>)

[Implementation](<https://github.com/mit-han-lab/temporal-shift-module>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DINOv3-L

Top-1 Accuracy (SS v2): 54. Result ID: 69b3dba847f34ce9249e5ef5.

[Reporting paper](<https://www.sota2.com/research/paper/69b39a370636e5bf4afbccff.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a931e209131cea83fc5.md>)

Recorded evaluation configuration: {"Model Size":"Large"}. This record does not assert independent reproduction.

### MoTIF-ST

Top-1 Accuracy (SS v2): 41.9. Result ID: 6a042eec47f34ce924e9e31f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

Recorded evaluation configuration: {"Backbone":"PE-G/14"}. This record does not assert independent reproduction.

### MoTIF

Top-1 Accuracy (SS v2): 40.4. Result ID: 6a042eec47f34ce924e9e31e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

Recorded evaluation configuration: {"Backbone":"PE-G/14"}. This record does not assert independent reproduction.

### MoTIF

Top-1 Accuracy (SS v2): 37.3. Result ID: 6a042eec47f34ce924e9e31d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

Recorded evaluation configuration: {"Backbone":"PE-L/14"}. This record does not assert independent reproduction.

### Global CBM

Top-1 Accuracy (SS v2): 33.6. Result ID: 6a042eec47f34ce924e9e31b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68f91e209131cea1f6a0.md>)

Recorded evaluation configuration: {"Backbone":"PE-G/14","Evaluation Protocol":"Supervised"}. This record does not assert independent reproduction.

### MoTIF

Top-1 Accuracy (SS v2): 25.8. Result ID: 6a042eec47f34ce924e9e31c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

Recorded evaluation configuration: {"Backbone":"ViT-L/14"}. This record does not assert independent reproduction.

### Zero-shot

Top-1 Accuracy (SS v2): 2.2. Result ID: 6a042eec47f34ce924e9e31a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7b991e209131cea89646.md>)

Recorded evaluation configuration: {"Backbone":"PE-G/14","Evaluation Protocol":"Zero-shot"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
