# Open-Vocabulary Audio-Visual Event Localization on AVEBench (OV seen)

Source: [Open-Vocabulary Audio-Visual Event Localization on AVEBench (OV seen)](<https://www.sota2.com/research/sota/open-vocabulary-audio-visual-event-localization-on-avebench-ov-seen>)

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

Data updated: 2026-02-26T05:30:21.827Z

Evaluation of Open-Vocabulary Audio-Visual Event Localization performance on the seen split of AVEBench.

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 | Open-Vocabulary Audio-Visual Event Localization |
| Dataset (including any stated split) | AVEBench OV (seen) |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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/open-vocabulary-audio-visual-event-localization-on-avebench-ov-seen.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [Segmentation Score](<https://www.sota2.com/research/sota/open-vocabulary-audio-visual-event-localization-on-avebench-ov-seen.md?metric=Segmentation%20Score>): Higher is better; unit: PERCENT.

- [Event Score](<https://www.sota2.com/research/sota/open-vocabulary-audio-visual-event-localization-on-avebench-ov-seen.md?metric=Event%20Score>): Higher is better; unit: PERCENT.

- [Average Score](<https://www.sota2.com/research/sota/open-vocabulary-audio-visual-event-localization-on-avebench-ov-seen.md?metric=Average%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| AVE | 76.6 | Not recorded | 68fa77581e209131cea72904 |
| MM-Pyramid | 76.5 | Not recorded | 68fa77581e209131cea72904 |
| PSP | 75.4 | Not recorded | 68fa77581e209131cea72904 |
| fine-tuning baseline | 72.5 | Not recorded | 68fa77581e209131cea72904 |
| CMRA | 65.2 | Not recorded | 68fa77581e209131cea72904 |
| training-free baseline | 57.5 | Not recorded | 68fa77581e209131cea72904 |
| CLIP & CLAP | 51.4 | Not recorded | 68fa77581e209131cea72904 |
| Video-LLaMA2 | 50.1 | Not recorded | 68fa77581e209131cea72904 |

### AVE

Accuracy: 76.6. Result ID: 699481673a6c3148491daef7.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa64271e209131cea02f3b.md>)

Recorded evaluation configuration: {"strategy":"fine-tuning"}. This record does not assert independent reproduction.

### MM-Pyramid

Accuracy: 76.5. Result ID: 699481673a6c3148491daef9.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6cf71e209131cea36e66.md>)

Recorded evaluation configuration: {"strategy":"fine-tuning"}. This record does not assert independent reproduction.

### PSP

Accuracy: 75.4. Result ID: 699481673a6c3148491daef8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69b41e209131cea23988.md>)

Recorded evaluation configuration: {"strategy":"fine-tuning"}. This record does not assert independent reproduction.

### fine-tuning baseline

Accuracy: 72.5. Result ID: 699481673a6c3148491daefa.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Implementation](<https://github.com/jasongief/OV-AVEL>)

Recorded evaluation configuration: {"strategy":"fine-tuning"}. This record does not assert independent reproduction.

### CMRA

Accuracy: 65.2. Result ID: 699481673a6c3148491daef6.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

Recorded evaluation configuration: {"strategy":"fine-tuning"}. This record does not assert independent reproduction.

### training-free baseline

Accuracy: 57.5. Result ID: 699481673a6c3148491daef5.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Implementation](<https://github.com/jasongief/OV-AVEL>)

Recorded evaluation configuration: {"strategy":"training-free"}. This record does not assert independent reproduction.

### CLIP & CLAP

Accuracy: 51.4. Result ID: 699481673a6c3148491daef4.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

Recorded evaluation configuration: {"strategy":"training-free"}. This record does not assert independent reproduction.

### Video-LLaMA2

Accuracy: 50.1. Result ID: 699481673a6c3148491daef3.

[Reporting paper](<https://www.sota2.com/research/paper/68fa77581e209131cea72904.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75761e209131cea681dd.md>)

[Implementation](<https://github.com/DAMO-NLP-SG/VideoLLaMA2>)

Recorded evaluation configuration: {"strategy":"training-free"}. This record does not assert independent reproduction.

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