Research map / Multi-agent systems and games / Games and equilibria
Generating Music Medleys via Playing Music Puzzle Games
Yu-Siang Huang, Szu-Yu Chou, Yi-Hsuan Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| learn convolutional features from frame-level similarity matrices | SEN versus competing models · Music jigsaw puzzle, n=3, fixed-length | global accuracy: 0.916 -> 0.994 | Improved |
| train SEN with downbeat-informed rather than uniform segmentation | SEN · Music sequencing | global accuracy: 0.440 -> 0.790 | Improved |
| train SEN with downbeat-informed rather than uniform segmentation | SEN · Music medley | global accuracy: 0.688 -> 0.750 | Improved |
| use cosine similarity instead of inner product for the similarity matrix (ablation) | SEN · Music sequencing | global accuracy: 0.38 -> 0.79 | Improved |
| keep convolution stride at 1 rather than increase it to 2 (ablation) | SEN · Music sequencing | global accuracy: 0.06 -> 0.79 | Improved |
| combine mean, max, and standard-deviation global pooling rather than use mean pooling alone (ablation) | SEN · Music sequencing | global accuracy: 0.49 -> 0.79 | Improved |
| combine mean, max, and standard-deviation global pooling rather than use max pooling alone (ablation) | SEN · Music sequencing | global accuracy: 0.76 -> 0.79 | Improved |
| use all three negative-pair types rather than R2R1 negatives only (ablation) | SEN · Music sequencing | global accuracy: 0.22 -> 0.79 | Improved |
| use all three negative-pair types rather than R1R3 negatives only (ablation) | SEN · Music sequencing | global accuracy: 0.46 -> 0.79 | Improved |
| use all three negative-pair types rather than R3R1 negatives only (ablation) | SEN · Music sequencing | global accuracy: 0.69 -> 0.79 | Improved |
Closest papers by meaning
- PerformanceNet: Score-to-Audio Music Generation with Multi-Band Convolutional Residual Network (AAAI 2019 · similarity 0.86)
- JEN-1 Composer: A Unified Framework for High-Fidelity Multi-Track Music Generation (AAAI 2025 · similarity 0.84)
- Whole-Song Hierarchical Generation of Symbolic Music Using Cascaded Diffusion Models (ICLR 2024 · similarity 0.84)
- Structured Multi-Track Accompaniment Arrangement via Style Prior Modelling (NeurIPS 2024 · similarity 0.83)
- DITTO: Diffusion Inference-Time T-Optimization for Music Generation (ICML 2024 · similarity 0.83)
- MusicRL: Aligning Music Generation to Human Preferences (ICML 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
