Pianist Transformer: Towards Expressive Piano Performance Rendering via Scalable Self-Supervised Pre-Training

ICLR 2026 Conference Submission17050 Authors

Keywords: Self-Supervised Learning, Music Performance Rendering, Model/Data Scaling

TL;DR: Pianist Transformer utilizes large-scale self-supervised pre-training on unlabeled MIDI data to achieve human-level expressive piano performance, overcoming the scaling limitations of previous supervised methods.

Abstract: Existing methods for expressive music performance rendering rely on supervised learning over small labeled datasets, which limits scaling of both data volume and model size, despite the availability of vast unlabeled music, as in vision and language. To address this gap, we introduce Pianist Transformer, with four key contributions: 1) a unified Musical Instrument Digital Interface (MIDI) data representation for learning the shared principles of musical structure and expression without explicit annotation; 2) an efficient asymmetric architecture, enabling longer contexts and faster inference without sacrificing rendering quality; 3) a self-supervised pre-training pipeline with 10B tokens and 135M-parameter model, unlocking data and model scaling advantages for expressive performance rendering; 4) a state-of-the-art performance model, which achieves strong objective metrics and human-level subjective ratings. Overall, Pianist Transformer establishes a scalable path toward human-like performance synthesis in music domain.

Primary Area: applications to computer vision, audio, language, and other modalities

Submission Number: 17050