Live Music Diffusion Models: Efficient Fine-Tuning and Post-Training of Interactive Diffusion Music Generators

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Novack, Zachary, Brade, Stephen, Kim, Haven, García, Hugo Flores, Shikarpur, Nithya, Talegaonkar, Chinmay, Kim, Suwan, Chen, Valerie K., McAuley, Julian, Berg-Kirkpatrick, Taylor, Huang, Cheng-Zhi Anna
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914588705947648
author Novack, Zachary
Brade, Stephen
Kim, Haven
García, Hugo Flores
Shikarpur, Nithya
Talegaonkar, Chinmay
Kim, Suwan
Chen, Valerie K.
McAuley, Julian
Berg-Kirkpatrick, Taylor
Huang, Cheng-Zhi Anna
author_facet Novack, Zachary
Brade, Stephen
Kim, Haven
García, Hugo Flores
Shikarpur, Nithya
Talegaonkar, Chinmay
Kim, Suwan
Chen, Valerie K.
McAuley, Julian
Berg-Kirkpatrick, Taylor
Huang, Cheng-Zhi Anna
contents Interactive streaming music generation promises the use of generative models for live performance and co-creation that is impossible with offline models. However, SOTA models exist in the discrete-AR regime, requiring industrial levels of compute for both training and inference. In this work, we investigate whether audio diffusion models, with their wide support in the open-source community but non-streaming bidirectional nature, can be repurposed efficiently into interactive models accessible on consumer hardware. By taking a critical look at the modern pipeline for block-wise outpainting diffusion, we identify critical inefficiencies during inference that result in strictly worse computational efficiency than their discrete-AR counterparts. We propose Live Music Diffusion Models (LMDMs), a simple modification of the generative diffusion process that recovers, and then outperforms, the inference complexity of the discrete Live Music Models (LMMs) through block-wise KV Caching. Unlike LMMs, LMDMs further enable stable post-training alignment through our novel ARC-Forcing paradigm, reducing error accumulation without any explicit RL or reward models. We demonstrate the application of LMDMs in a number of creative domains, including text-conditioned generation, sketch-based music synthesis, and jamming. We finally show how LMDMs can be used as a generative instrument in a real artist-AI collaboration, utilizing LMDMs as a "generative delay" to transform musicians' improvisation live for variable timbral effects while running locally on a consumer gaming laptop.
format Preprint
id arxiv_https___arxiv_org_abs_2605_22717
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Live Music Diffusion Models: Efficient Fine-Tuning and Post-Training of Interactive Diffusion Music Generators
Novack, Zachary
Brade, Stephen
Kim, Haven
García, Hugo Flores
Shikarpur, Nithya
Talegaonkar, Chinmay
Kim, Suwan
Chen, Valerie K.
McAuley, Julian
Berg-Kirkpatrick, Taylor
Huang, Cheng-Zhi Anna
Sound
Artificial Intelligence
Machine Learning
Multimedia
Interactive streaming music generation promises the use of generative models for live performance and co-creation that is impossible with offline models. However, SOTA models exist in the discrete-AR regime, requiring industrial levels of compute for both training and inference. In this work, we investigate whether audio diffusion models, with their wide support in the open-source community but non-streaming bidirectional nature, can be repurposed efficiently into interactive models accessible on consumer hardware. By taking a critical look at the modern pipeline for block-wise outpainting diffusion, we identify critical inefficiencies during inference that result in strictly worse computational efficiency than their discrete-AR counterparts. We propose Live Music Diffusion Models (LMDMs), a simple modification of the generative diffusion process that recovers, and then outperforms, the inference complexity of the discrete Live Music Models (LMMs) through block-wise KV Caching. Unlike LMMs, LMDMs further enable stable post-training alignment through our novel ARC-Forcing paradigm, reducing error accumulation without any explicit RL or reward models. We demonstrate the application of LMDMs in a number of creative domains, including text-conditioned generation, sketch-based music synthesis, and jamming. We finally show how LMDMs can be used as a generative instrument in a real artist-AI collaboration, utilizing LMDMs as a "generative delay" to transform musicians' improvisation live for variable timbral effects while running locally on a consumer gaming laptop.
title Live Music Diffusion Models: Efficient Fine-Tuning and Post-Training of Interactive Diffusion Music Generators
topic Sound
Artificial Intelligence
Machine Learning
Multimedia
url https://arxiv.org/abs/2605.22717