# Audio-visual Recognition on EPIC-SOUNDS (test val)

Source: [Audio-visual Recognition on EPIC-SOUNDS (test val)](<https://www.sota2.com/research/sota/audio-visual-recognition-on-epic-sounds-test-val>)

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

Data updated: 2026-02-26T05:30:26.876Z

Evaluation of Audio-visual Recognition performance on the test/validation split of the EPIC-SOUNDS dataset, measured using Accuracy.

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 | Audio-visual Recognition |
| Dataset (including any stated split) | EPIC-SOUNDS (test val) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 5 of 5 |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/audio-visual-recognition-on-epic-sounds-test-val.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| CA2ST | 61 | Not recorded | 68fa77ad1e209131cea74a67 |
| CAVA | 60.3 | Not recorded | 68fa77ad1e209131cea74a67 |
| CAST | 47.8 | Not recorded | 68fa77ad1e209131cea74a67 |
| CLIP-B | 31.9 | Not recorded | 68fa77ad1e209131cea74a67 |
| VideoMAE-B | 31.8 | Not recorded | 68fa77ad1e209131cea74a67 |

### CA2ST

Accuracy: 61. Result ID: 69955bd43badfaffe64d59b8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

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

Recorded evaluation configuration: {"Param.":"320M","Throughput (V/s)":30,"Latency (ms)":54.8,"GFLOPs":454}. This record does not assert independent reproduction.

### CAVA

Accuracy: 60.3. Result ID: 69955bd43badfaffe64d59b7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

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

Recorded evaluation configuration: {"Param.":"216M","Throughput (V/s)":65,"Latency (ms)":36.6,"GFLOPs":236}. This record does not assert independent reproduction.

### CAST

Accuracy: 47.8. Result ID: 69955bd43badfaffe64d59b6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

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

Recorded evaluation configuration: {"Param.":"217M","Throughput (V/s)":37,"Latency (ms)":38.5,"GFLOPs":391}. This record does not assert independent reproduction.

### CLIP-B

Accuracy: 31.9. Result ID: 69955bd43badfaffe64d59b5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Param.":"86M","Throughput (V/s)":81,"Latency (ms)":14.7,"GFLOPs":152}. This record does not assert independent reproduction.

### VideoMAE-B

Accuracy: 31.8. Result ID: 69955bd43badfaffe64d59b4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77ad1e209131cea74a67.md>)

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

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

Recorded evaluation configuration: {"Param.":"87M","Throughput (V/s)":67,"Latency (ms)":17.6,"GFLOPs":192}. This record does not assert independent reproduction.

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