# Audio Question Answering on speech-chatbot-alpaca-eval (S-Alpaca)

Source: [Audio Question Answering on speech-chatbot-alpaca-eval (S-Alpaca)](<https://www.sota2.com/research/sota/audio-question-answering-on-speech-chatbot-alpaca-eval-s-alpaca>)

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

Data updated: 2026-04-10T00:08:12.000Z

Evaluation of model performance on the speech-chatbot-alpaca-eval (S-Alpaca) dataset for Audio Question Answering, 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 Question Answering |
| Dataset (including any stated split) | speech-chatbot-alpaca-eval (S-Alpaca) |
| 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/audio-question-answering-on-speech-chatbot-alpaca-eval-s-alpaca.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| AudioKV | 66.7 | Not recorded | 69d732b37d0172140ec76c87 |
| PyramidKV | 65.2 | Not recorded | 69d732b37d0172140ec76c87 |
| SnapKV | 64.7 | Not recorded | 69d732b37d0172140ec76c87 |
| AdaKV | 64.7 | Not recorded | 69d732b37d0172140ec76c87 |
| AudioKV | 57.6 | Not recorded | 69d732b37d0172140ec76c87 |
| AdaKV | 56.1 | Not recorded | 69d732b37d0172140ec76c87 |
| PyramidKV | 56.1 | Not recorded | 69d732b37d0172140ec76c87 |
| SnapKV | 54.5 | Not recorded | 69d732b37d0172140ec76c87 |

### AudioKV

Accuracy: 66.7. Result ID: 69d83ec647f34ce924c03619.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-7B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

### PyramidKV

Accuracy: 65.2. Result ID: 69d83ec647f34ce924c03618.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78841e209131cea78cba.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-7B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

### SnapKV

Accuracy: 64.7. Result ID: 69d83ec647f34ce924c03616.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df89.md>)

[Implementation](<https://github.com/FasterDecoding/SnapKV>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-7B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

### AdaKV

Accuracy: 64.7. Result ID: 69d83ec647f34ce924c03617.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c2e1e209131cea8ce46.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-7B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

### AudioKV

Accuracy: 57.6. Result ID: 69d83ec647f34ce924c03615.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-3B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

### AdaKV

Accuracy: 56.1. Result ID: 69d83ec647f34ce924c03613.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c2e1e209131cea8ce46.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-3B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

### PyramidKV

Accuracy: 56.1. Result ID: 69d83ec647f34ce924c03614.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78841e209131cea78cba.md>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-3B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

### SnapKV

Accuracy: 54.5. Result ID: 69d83ec647f34ce924c03612.

[Reporting paper](<https://www.sota2.com/research/paper/69d732b37d0172140ec76c87.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73b21e209131cea5df89.md>)

[Implementation](<https://github.com/FasterDecoding/SnapKV>)

Recorded evaluation configuration: {"Model":"Qwen2.5-Omni-3B","KV cache budget":"40%","max\_gen\_tokens":"64","Judge":"Gemini 2.5 Flash"}. This record does not assert independent reproduction.

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