---
title: "QA-MDT: Quality-aware Masked Diffusion Transformer for Enhanced Music Generation"
authors:
  - "Chang Li"
  - "Ruoyu Wang"
  - "Lijuan Liu"
  - "Jun Du"
  - "Yixuan Sun"
  - "Zilu Guo"
  - "Zhengrong Zhang"
  - "Yuan Jiang"
  - "Jianqing Gao"
  - "Feng Ma"
doi: "10.24963/ijcai.2025/1126"
url: "https://ijcai.org/proceedings/2025/1126"
pdf: "https://ijcai.org/proceedings/2025/1126.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "10135-10143"
track: "AI, Arts & Creativity"
keywords:
  - "Application domains: General"
  - "Application domains: Music and sound"
  - "Theory and philosophy of arts and creativity in AI systems: Autonomous creative or artistic AI"
  - "Methods and resources: Machine learning, deep learning, neural models, reinforcement learning"
---

# QA-MDT: Quality-aware Masked Diffusion Transformer for Enhanced Music Generation

Chang Li, Ruoyu Wang, Lijuan Liu, Jun Du, Yixuan Sun, Zilu Guo, Zhengrong Zhang, Yuan Jiang, Jianqing Gao, Feng Ma

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 10135-10143. DOI: [10.24963/ijcai.2025/1126](https://doi.org/10.24963/ijcai.2025/1126)

## Abstract

Text-to-music (TTM) generation, which converts textual descriptions into audio, opens up innovative avenues for multimedia creation.
Achieving high quality and diversity in this process demands extensive, high-quality data, which are often scarce in available datasets. Most open-source datasets frequently suffer from issues like low-quality waveforms and low text-audio consistency, hindering the advancement of music generation models.
To address these challenges, we propose a novel quality-aware training paradigm for generating high-quality, high-musicality music from large-scale, quality-imbalanced datasets. Additionally, by leveraging unique properties in the latent space of musical signals, we adapt and implement a masked diffusion transformer (MDT) model for the TTM task, showcasing its capacity for quality control and enhanced musicality. Furthermore, we introduce a three-stage caption refinement approach to address low-quality captions' issue. Experiments show state-of-the-art (SOTA) performance on benchmark datasets including MusicCaps and the Song-Describer Dataset with both objective and subjective metrics.
Demo audio samples are available at https://qa-mdt.github.io/, code and pretrained checkpoints are open-sourced at https://github.com/ivcylc/OpenMusic.