Pith. sign in

REVIEW 3 cited by

Frequency-Aware Masked Autoencoders for Multimodal Pretraining on Biosignals

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2309.05927 v2 pith:7MAOPKNF submitted 2023-09-12 cs.LG cs.AIeess.SP

classification cs.LGcs.AIeess.SP
keywords pretrainingbiosignalsmultimodalfrequency-awaremaskedmodalityarchitecturedistributional
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Leveraging multimodal information from biosignals is vital for building a comprehensive representation of people's physical and mental states. However, multimodal biosignals often exhibit substantial distributional shifts between pretraining and inference datasets, stemming from changes in task specification or variations in modality compositions. To achieve effective pretraining in the presence of potential distributional shifts, we propose a frequency-aware masked autoencoder ($\texttt{bio}$FAME) that learns to parameterize the representation of biosignals in the frequency space. $\texttt{bio}$FAME incorporates a frequency-aware transformer, which leverages a fixed-size Fourier-based operator for global token mixing, independent of the length and sampling rate of inputs. To maintain the frequency components within each input channel, we further employ a frequency-maintain pretraining strategy that performs masked autoencoding in the latent space. The resulting architecture effectively utilizes multimodal information during pretraining, and can be seamlessly adapted to diverse tasks and modalities at test time, regardless of input size and order. We evaluated our approach on a diverse set of transfer experiments on unimodal time series, achieving an average of $\uparrow$5.5% improvement in classification accuracy over the previous state-of-the-art. Furthermore, we demonstrated that our architecture is robust in modality mismatch scenarios, including unpredicted modality dropout or substitution, proving its practical utility in real-world applications. Code is available at https://github.com/apple/ml-famae .

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A robust PPG foundation model using multimodal physiological supervision

    cs.LG 2026-06 unverdicted novelty 6.0 of 10

    A PPG foundation model pretrained via multimodal ECG/respiratory contrastive sample selection on ICU data improves performance on 14 of 15 downstream tasks including field-like data while using 3x fewer subjects.

  2. PAMF: Prior-Aware Multimodal Fusion for Incomplete Time Series Data

    cs.LG 2026-06 unverdicted novelty 6.0 of 10

    PAMF initializes flow matching with missingness-type priors and shares encoder weights between imputation and classification to improve multimodal time-series prediction under incomplete observations.

  3. Event Fields: Learning Latent Event Structure for Waveform Foundation Models

    cs.LG 2026-05 unverdicted novelty 6.0 of 10

    Event-centric waveform foundation models are learned via self-supervised consistency on latent event structures and interactions, yielding improved performance and label efficiency over sequence-based baselines on phy...

Pith tools