AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

AudioMosaic: Contrastive Masked Audio Representation Learning

Hanxun Huang, Qizhou Wang, Xingjun Ma, Cihang Xie, Christopher Leckie, Sarah Monazam Erfani

ICML 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Contrastive versus non-contrastive theory · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use contrastive time-frequency masked pre-training instead of Audio-MAE reconstructionViT-B/16 audio encoder · AS-20K linear probingmAP: 18.3 -> 29.4Improved
AudioMosaic vs SSLAM under fine-tuningAudioMosaic · AS-20KmAP: 40.9 -> 42.5Improved
AudioMosaic vs SSLAM under fine-tuningAudioMosaic · AS-2MmAP: 50.2 -> 50.2No clear effect
aggregate layers with attentive probe versus best single layer (ablation)AudioMosaic linear probe · AS-20KmAP: 30.2 -> 33.5Improved
time-frequency masking at inference instead of no masking when measuring effective rank (ablation)encoders · effective rank analysiseffective rankImproved
use time-frequency masking instead of unstructured masking for contrastive positive pairs (ablation)AudioMosaic · AS-20K fine-tuningmAPImproved
add frequency masking on top of time masking at high mask ratios (ablation)AudioMosaic · AS-20K fine-tuningmAPImproved
time-only masking vs time-frequency at low mask ratios (ablation)AudioMosaic · AS-20K fine-tuningmAPNo clear effect
increase pre-training batch size up to 6144 (ablation)AudioMosaic · AS-20K fine-tuningmAPImproved
use rho_t 0.6 and rho_f 0.4 masking ratios (ablation)AudioMosaic · AS-20K fine-tuningmAP: 39.0 best cell in gridImproved
use AASIST head instead of linear head on AudioMosaic (ablation)AudioMosaic · EnvSDD TTA averageEER (%): 1.30 -> 1.41No clear effect
replace CAV-MAE encoder with AudioMosaic in LTULTU with LLaMA-7B · VGGSoundaccuracy: 38.4 -> 54.6Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy