Research map / Probabilistic and scientific modeling / Variational and generative sampling
A Hierarchical Latent Vector Model for Learning Long-Term Structure in Music
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE · 16-bar Melody | sampling reconstruction accuracy: 0.620 -> 0.812 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE · 16-bar Drum | sampling reconstruction accuracy: 0.549 -> 0.879 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE · Trio (Melody) | sampling reconstruction accuracy: 0.579 -> 0.753 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE · Trio (Bass) | sampling reconstruction accuracy: 0.565 -> 0.773 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE · Trio (Drums) | sampling reconstruction accuracy: 0.641 -> 0.863 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE trained on Lakh MIDI · Lakh MIDI 16-bar Melody | sampling reconstruction accuracy: 0.685 -> 0.867 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE trained on Lakh MIDI · Lakh MIDI 16-bar Drum | sampling reconstruction accuracy: 0.794 -> 0.908 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE trained on Lakh MIDI · Lakh MIDI Trio (Melody) | sampling reconstruction accuracy: 0.660 -> 0.760 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE trained on Lakh MIDI · Lakh MIDI Trio (Bass) | sampling reconstruction accuracy: 0.651 -> 0.782 | Helped |
| Replace the flat decoder with a hierarchical conductor and subsequence decoders | MusicVAE trained on Lakh MIDI · Lakh MIDI Trio (Drums) | sampling reconstruction accuracy: 0.641 -> 0.895 | Helped |
| Use an autoregressive conductor that receives decoder state between subsequences (ablation) | MusicVAE hierarchical decoder · Music modeling | performance | Hurt |
Closest papers by meaning
- Compound Word Transformer: Learning to Compose Full-Song Music over Dynamic Directed Hypergraphs (AAAI 2021 · similarity 0.89)
- Discovering and Steering Interpretable Concepts in Large Generative Music Models (ICLR 2026 · similarity 0.88)
- MusER: Musical Element-Based Regularization for Generating Symbolic Music with Emotion (AAAI 2024 · similarity 0.88)
- MoMu-Diffusion: On Learning Long-Term Motion-Music Synchronization and Correspondence (NeurIPS 2024 · similarity 0.87)
- Whole-Song Hierarchical Generation of Symbolic Music Using Cascaded Diffusion Models (ICLR 2024 · similarity 0.87)
- From Isolated feature to Orbits:\\Discovering Music Concepts via Multi-SAE Alignment (NeurIPS 2026 · similarity 0.86)
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.