# Generalized Zero-Shot Retrieval (Text-to-Audio-Video) on AudioSet ZSL (test)

Source: [Generalized Zero-Shot Retrieval (Text-to-Audio-Video) on AudioSet ZSL (test)](<https://www.sota2.com/research/sota/generalized-zero-shot-retrieval-text-to-audio-video-on-audioset-zsl-test>)

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

Data updated: 2026-02-26T06:09:52.886Z

Evaluation of Generalized Zero-Shot Retrieval performance, specifically Text-to-Audio-Video, on the test split of the AudioSet ZSL dataset, measured using mAP for Seen (S), Unseen (U), and Harmonic Mean (HM) categories.

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 | Generalized Zero-Shot Retrieval (Text-to-Audio-Video) |
| Dataset (including any stated split) | AudioSet ZSL (test) |
| Metric | mAP (Seen) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [mAP (Seen)](<https://www.sota2.com/research/sota/generalized-zero-shot-retrieval-text-to-audio-video-on-audioset-zsl-test.md?metric=mAP%20(Seen)>): Higher is better; unit: PERCENT.

- [mAP (Unseen)](<https://www.sota2.com/research/sota/generalized-zero-shot-retrieval-text-to-audio-video-on-audioset-zsl-test.md?metric=mAP%20(Unseen)>): Higher is better; unit: PERCENT.

- [mAP (HM)](<https://www.sota2.com/research/sota/generalized-zero-shot-retrieval-text-to-audio-video-on-audioset-zsl-test.md?metric=mAP%20(HM)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | mAP (Seen) | Result date | Report paper |
| --- | --- | --- | --- |
| AVGZSLNet | 72.25 | Not recorded | 68fa68c11e209131cea1e112 |
| AVGZSLNet | 68.94 | Not recorded | 68fa68c11e209131cea1e112 |
| CJME | 65.45 | Not recorded | 68fa68c11e209131cea1e112 |
| CJME | 62.97 | Not recorded | 68fa68c11e209131cea1e112 |

### AVGZSLNet

mAP (Seen): 72.25. Result ID: 698d82e18b517eb433220cde.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68c11e209131cea1e112.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68c11e209131cea1e112.md>)

[Implementation](<https://github.com/krantiparida/AudioSetZSL>)

Recorded evaluation configuration: {"Test":"T -\> AV","Fusion":"equal weight"}. This record does not assert independent reproduction.

### AVGZSLNet

mAP (Seen): 68.94. Result ID: 698d82e18b517eb433220ce0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68c11e209131cea1e112.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa68c11e209131cea1e112.md>)

[Implementation](<https://github.com/krantiparida/AudioSetZSL>)

Recorded evaluation configuration: {"Test":"T -\> AV","Fusion":"attention"}. This record does not assert independent reproduction.

### CJME

mAP (Seen): 65.45. Result ID: 698d82e18b517eb433220cdd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68c11e209131cea1e112.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66931e209131cea1106d.md>)

[Implementation](<https://www.cse.iitk.ac.in/users/kranti/avzsl.html>)

Recorded evaluation configuration: {"Test":"T -\> AV","Fusion":"equal weight"}. This record does not assert independent reproduction.

### CJME

mAP (Seen): 62.97. Result ID: 698d82e18b517eb433220cdf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa68c11e209131cea1e112.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa66931e209131cea1106d.md>)

[Implementation](<https://www.cse.iitk.ac.in/users/kranti/avzsl.html>)

Recorded evaluation configuration: {"Test":"T -\> AV","Fusion":"attention"}. This record does not assert independent reproduction.

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