S-SONDO distills general audio foundation models into students up to 61 times smaller while retaining up to 96% of teacher performance using only output embeddings.
Matpac++: Enhanced masked latent predic- tion for self-supervised audio representation learning
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
abstract
Masked latent prediction has emerged as a leading paradigm in self-supervised learning (SSL), especially for general audio and music representation learning. While recent methods have demonstrated strong performance, the role of the predictor module used at the output of such SSL systems remains mainly overlooked, despite being crucial for solving the pretext task at hand. In particular, this module should be able to deal with the ambiguity inherent in audio content, especially when it is composed of multiple sound sources. This work proposes a novel enhancement: integrating Multiple Choice Learning (MCL) to explicitly model prediction ambiguity and improve representation quality. We build on top of the recently proposed MATPAC system, improving its prediction and unsupervised classification pretext tasks with MCL. We extensively evaluate our method, MATPAC++, through both linear probing across multiple downstream tasks and fine-tuning on AudioSet, employing a unified protocol that enables rigorous and fair comparisons with state-of-the-art SSL approaches. Results show that our proposal achieves state-of-the-art when fine-tuned on AudioSet and overall state-of-the-art scores on downstream tasks. Additionally, we examine domain specialisation by training exclusively on music data, where our model achieves state-of-the-art performance with significantly improved efficiency.
years
2026 4representative citing papers
Event-level MAE embeddings plus UMAP/HDBSCAN or K-Means clustering recover 15 hydroacoustic classes from multi-year Mayotte data with ~1 hour of annotation and detector-comparable F1.
PupuJEPA applies a visual JEPA framework to 2D spectrograms with music-specific adaptations and outperforms 1D SSL models on the MARBLE benchmark for multiple MIR tasks.
TinyMU is a 229M-parameter compact music understanding model that achieves 82% of state-of-the-art large audio-language model performance on the MuChoMusic benchmark while being 35 times smaller.
citing papers explorer
-
S-SONDO: Self-Supervised Knowledge Distillation for General Audio Foundation Models
S-SONDO distills general audio foundation models into students up to 61 times smaller while retaining up to 96% of teacher performance using only output embeddings.
-
A Self-Supervised Approach for Minimal-Annotation Hydroacoustic Data Exploration
Event-level MAE embeddings plus UMAP/HDBSCAN or K-Means clustering recover 15 hydroacoustic classes from multi-year Mayotte data with ~1 hour of annotation and detector-comparable F1.
-
Frequency-Aware Self-Supervised Music Representation Learning
PupuJEPA applies a visual JEPA framework to 2D spectrograms with music-specific adaptations and outperforms 1D SSL models on the MARBLE benchmark for multiple MIR tasks.
-
TinyMU: A Compact Audio-Language Model for Music Understanding
TinyMU is a 229M-parameter compact music understanding model that achieves 82% of state-of-the-art large audio-language model performance on the MuChoMusic benchmark while being 35 times smaller.