Read-only mirror of the MIREX Wiki, snapshot of 29 October 2026.
Open this page on the original wiki to edit it or to see newer changes.
2025:Music Description & Captioning Results
From MIREX Wiki
Results[edit]
| System | Methods Used | Trained on the training set of | ROUGE-L | BLEU (Avg. 1-4) | METEOR |
|---|---|---|---|---|---|
| Baseline 1 | LP-MusicCaps | LP-MusicCaps Dataset | 0.119 | 0.033 | 0.167 |
| Baseline 2 | MusiLingo | MusicInstruct Dataset | 0.302* | 0.081* | 0.143* |
| Baseline 3 | Qwen2-Audio | Multiple datasets | 0.285* | 0.234* | 0.285* |
- Reported by paper results.