# Integrating Contextual Embeddings into Evaluation of Expressive MIDI Piano Performances

arXiv: https://arxiv.org/abs/2607.27909 · HTML: https://mlsanity.com/paper/2607.27909/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 36 references that are in the map.
- First posted on arXiv 30 Jul 2026.

## Builds on (most cited first, 26 of 26)

- [A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks](https://mlsanity.com/paper/1807.03888/) (2018)
- [Demystifying MMD GANs](https://mlsanity.com/paper/1801.01401/) (2018)
- [Rethinking FID: Towards a Better Evaluation Metric for Image Generation](https://mlsanity.com/paper/2401.09603/) (2023)
- [A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection](https://mlsanity.com/paper/2106.09022/) (2021)
- [Effectively Unbiased FID and Inception Score and where to find them](https://mlsanity.com/paper/1911.07023/) (2019)
- [Adapting Frechet Audio Distance for Generative Music Evaluation](https://mlsanity.com/paper/2311.01616/) (2023)
- [Gibbs Sampling with People](https://mlsanity.com/paper/2008.02595/) (2020)
- [Conditional Generative Moment-Matching Networks](https://mlsanity.com/paper/1606.04218/) (2016)
- [KAD: No More FAD! An Effective and Efficient Evaluation Metric for Audio Generation](https://mlsanity.com/paper/2502.15602/) (2025)
- [CLaMP 3: Universal Music Information Retrieval Across Unaligned Modalities and Unseen Languages](https://mlsanity.com/paper/2502.10362/) (2025)
- [Aligning Text-to-Music Evaluation with Human Preferences](https://mlsanity.com/paper/2503.16669/) (2025)
- [PDMX: A Large-Scale Public Domain MusicXML Dataset for Symbolic Music Processing](https://mlsanity.com/paper/2409.10831/) (2024)
- [Frechet Music Distance: A Metric For Generative Symbolic Music Evaluation](https://mlsanity.com/paper/2412.07948/) (2024)
- [CLaMP 2: Multimodal Music Information Retrieval Across 101 Languages Using Large Language Models](https://mlsanity.com/paper/2410.13267/) (2024)
- [DExter: Learning and Controlling Performance Expression with Diffusion Models](https://mlsanity.com/paper/2406.14850/) (2024)
- [Evaluating Aleatoric Uncertainty via Conditional Generative Models](https://mlsanity.com/paper/2206.04287/) (2022)
- [Aria-MIDI: A Dataset of Piano MIDI Files for Symbolic Music Modeling](https://mlsanity.com/paper/2504.15071/) (2025)
- [Scaling Self-Supervised Representation Learning for Symbolic Piano Performance](https://mlsanity.com/paper/2506.23869/) (2025)
- [Reconstructing Human Expressiveness in Piano Performances with a Transformer Network](https://mlsanity.com/paper/2306.06040/) (2023)
- [Online Symbolic Music Alignment with Offline Reinforcement Learning](https://mlsanity.com/paper/2401.00466/) (2023)
- [LLaQo: Towards a Query-Based Coach in Expressive Music Performance Assessment](https://mlsanity.com/paper/2409.08795/) (2024)
- [Towards An Integrated Approach for Expressive Piano Performance Synthesis from Music Scores](https://mlsanity.com/paper/2501.10222/) (2025)
- [From Audio Encoders to Piano Judges: Benchmarking Performance Understanding for Solo Piano](https://mlsanity.com/paper/2407.04518/) (2024)
- [From Aesthetics to Human Preferences: Comparative Perspectives of Evaluating Text-to-Music Systems](https://mlsanity.com/paper/2504.21815/) (2025)
- [SyMuPe: Affective and Controllable Symbolic Music Performance](https://mlsanity.com/paper/2511.03425/) (2025)
- [Sounding Out Reconstruction Error-Based Evaluation of Generative Models of Expressive Performance](https://mlsanity.com/paper/2401.00471/) (2023)

## Abstract (arXiv)

Objective evaluation of expressive MIDI piano performances typically relies on attribute statistics such as timing, velocity, and duration of individual notes. However, these methods often disregard dependencies between notes, which poses a potential limitation in assessing the similarity between two sets of performances. In generative applications, the wide variety of expressive attributes makes it difficult to aggregate them into a single scalar metric for model selection. In this work, we reexamine attribute-scoped metrics and explore the perceptual properties of contextual embeddings from self-supervised symbolic music models, Aria and CLaMP3. Results from our listening study indicate that these models can be used as perceptual proxies, showing agreement with per-sample human ratings on par with traditional metrics. To measure conditional distributional similarity, we adapt Kernel Audio Distance to the symbolic music domain. Unlike Pearson correlation and reconstruction error, kernel-based methods on contextual embeddings do not require note alignment and are sensitive to contextual perturbations. To facilitate reproducibility, we release Pereval, an open-source library that integrates performance evaluation utilities, including both attribute-scoped and deep feature metrics.