# Question Answering on MuSiQuest (F1 score, ROUGE-L)

Source: [Question Answering on MuSiQuest (F1 score, ROUGE-L)](<https://www.sota2.com/research/sota/question-answering-on-musiquest-f1-score-rouge-l>)

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

Data updated: 2026-02-26T05:43:42.722Z

Evaluation of model performance on the MuSiQuest dataset for Question Answering tasks, measured using F1 score and ROUGE-L.

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 | Question Answering |
| Dataset (including any stated split) | MuSiQuest |
| Metric | F1 Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 1 of 1 |
| 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

- [F1 Score](<https://www.sota2.com/research/sota/question-answering-on-musiquest-f1-score-rouge-l.md?metric=F1%20Score>): Higher is better; unit: SCALAR.

- [ROUGE-L](<https://www.sota2.com/research/sota/question-answering-on-musiquest-f1-score-rouge-l.md?metric=ROUGE-L>): Higher is better; unit: SCALAR.

## Recorded results

| Method | F1 Score | Result date | Report paper |
| --- | --- | --- | --- |
| KVComm | 0.36 | Not recorded | 68fa7be81e209131cea8b3ea |

### KVComm

F1 Score: 0.36. Result ID: 699a3f9d5698665ec4c656a8.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7be81e209131cea8b3ea.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7be81e209131cea8b3ea.md>)

Recorded evaluation configuration: {"Sender Model (Ms)":"huihui-ai/Llama-3.2-3B-Instruct-abliterated","Receiver Model (Mr)":"suayptalha/DeepSeek-R1-Distill-Llama-3B","Communication Ratio (M)":0.7}. This record does not assert independent reproduction.

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