# Generative Modeling of Bach-Style Symbolic Music: A Comparative Study of Autoregressive, Latent-Variable, and Adversarial Approaches

arXiv: https://arxiv.org/abs/2606.13626 · HTML: https://mlsanity.com/paper/2606.13626/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 15 references that are in the map.
- First posted on arXiv 11 Jun 2026.

## Builds on (most cited first, 8 of 8)

- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [A Hierarchical Latent Vector Model for Learning Long-Term Structure in Music](https://mlsanity.com/paper/1803.05428/) (2018)
- [Conditional LSTM-GAN for Melody Generation from Lyrics](https://mlsanity.com/paper/1908.05551/) (2019)
- [Self-Supervised VQ-VAE for One-Shot Music Style Transfer](https://mlsanity.com/paper/2102.05749/) (2021)
- [Bach Style Music Authoring System based on Deep Learning](https://mlsanity.com/paper/2110.02640/) (2021)

## Abstract (arXiv)

We study generative modeling of Bach-style symbolic piano music using a shared MIDI corpus and three model families: autoregressive LSTMs with attention, latent-variable models including recurrent VAEs and vector-quantized VAEs, and generative adversarial networks. We compare their ability to model polyphonic note sequences, learn useful latent representations, and generate stylistically coherent compositions. Our experiments show that the autoregressive LSTM with attention produces the most musically coherent samples, while vector quantization helps mitigate posterior collapse and yields more structured outputs than conventional recurrent VAEs. The adversarial approach captures local pitch patterns but remains difficult to train and generalizes less reliably to Bach's style. These results highlight the relative strengths and failure modes of autoregressive, latent-variable, and adversarial approaches for symbolic music generation.