Research map / Image generation / Autoregressive and latent generation
ComMU: Dataset for Combinatorial Music Generation
Lee Hyun, Taehyun Kim, Hyolim Kang, Minjoo Ki, Hyeonchan Hwang, Kwanho Park, Sharang Han, Seon Joo Kim
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 |
|---|---|---|---|
| increase sampling temperature from 0.7 to 1.2 (ablation) | Transformer-XL on ComMU · ComMU validation | diversity (D): 0.2626 -> 0.3688 | Improved |
| increase sampling temperature from 0.7 to 1.2 (ablation) | Transformer-XL on ComMU · ComMU validation | pitch controllability (CP): 0.8798 -> 0.7721 | Hurt |
| increase sampling temperature from 0.7 to 1.2 (ablation) | Transformer-XL on ComMU · ComMU validation | velocity controllability (CV): 0.9696 -> 0.8566 | Hurt |
| increase top-k from 32 to 200 at τ=0.95 (ablation) | Transformer-XL on ComMU · ComMU validation | pitch controllability (CP): 0.8412 -> 0.8488 | No clear effect |
| include key and BPM augmentation rather than no augmentation (ablation) | Transformer-XL on ComMU · ComMU validation | velocity controllability (CV): 0.8194 -> 0.9102 | Improved |
| include key and BPM augmentation rather than no augmentation (ablation) | Transformer-XL on ComMU · ComMU validation | diversity (D): 0.4386 -> 0.3160 | Hurt |
| use key augmentation only instead of key and BPM augmentation (ablation) | Transformer-XL on ComMU · ComMU validation | velocity controllability (CV): 0.9102 -> 0.8663 | Hurt |
| use BPM augmentation only instead of key and BPM augmentation (ablation) | Transformer-XL on ComMU · ComMU validation | velocity controllability (CV): 0.9102 -> 0.7615 | Hurt |
| use Transformer-GAN rather than Transformer-XL (ablation) | ComMU generation · ComMU validation | diversity (D): 0.3160 -> 0.4097 | Improved |
| use Transformer-GAN rather than Transformer-XL (ablation) | ComMU generation · ComMU validation | velocity controllability (CV): 0.9102 -> 0.8699 | Hurt |
| use REMI-style representation rather than MIDI-like representation (ablation) | Transformer-XL · Subjective evaluation | controllability: 2.53 -> 9.49 | Improved |
| increase representation resolution from 32 to 128 (ablation) | Transformer-XL with REMI · Subjective evaluation | overall: 6.99 -> 7.31 | Improved |
Closest papers by meaning
- Aria-MIDI: A Dataset of Piano MIDI Files for Symbolic Music Modeling (ICLR 2025 · similarity 0.87)
- MuPT: A Generative Symbolic Music Pretrained Transformer (ICLR 2025 · similarity 0.86)
- Compose with Me: Collaborative Music Inpainter for Symbolic Music Infilling (AAAI 2025 · similarity 0.86)
- Efficient Fine-Grained Guidance for Diffusion Model Based Symbolic Music Generation (ICML 2025 · similarity 0.86)
- Unifying Symbolic Music Arrangement: Track-Aware Reconstruction and Structured Tokenization (NeurIPS 2025 · similarity 0.85)
- Whole-Song Hierarchical Generation of Symbolic Music Using Cascaded Diffusion Models (ICLR 2024 · similarity 0.85)
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.
