PaperarXiv:2509.19215
PPG-Distill: Efficient Photoplethysmography Signals Analysis via Foundation Model Distillation
NeurIPS 2025 Workshop on Learning from Time Series for Health (TS4H Spotlight)
TL;DR
PPG-Distill distills large PPG foundation models into small students with prediction, feature, morphology, and rhythm losses. On DaLiA and StanfordAF a GPT-PPG-1m student recovers most teacher accuracy while cutting compute and memory for on-device use.
- NeurIPS 2025 TS4H Spotlight
- arXiv 2509.19215
- Emory CS + Nell Hodgson Woodruff School of Nursing
- DaLiA HR MAE 8.34; StanfordAF F1 0.77
- Open training code; no packaged FM weights
Results on ChatPPG leaderboards
All leaderboardsNo results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.
Tasks & signals
Signals
Devices
Keywords
Datasets
Models
How it works
Cite this paper
APA
Ni, J., Kataria, S., Tang, S., Yang, C., Hu, X., & Jin, W. (2025). PPG-Distill: Efficient Photoplethysmography Signals Analysis via Foundation Model Distillation. NeurIPS 2025 Workshop on Learning from Time Series for Health. https://arxiv.org/abs/2509.19215
BibTeX
@inproceedings{ni2025ppgdistill,
title = {{PPG-Distill: Efficient Photoplethysmography Signals Analysis via Foundation Model Distillation}},
author = {Ni, Juntong and Kataria, Saurabh and Tang, Shengpu and Yang, Carl and Hu, Xiao and Jin, Wei},
year = {2025},
booktitle = {NeurIPS 2025 Workshop on Learning from Time Series for Health},
eprint = {2509.19215},
archivePrefix = {arXiv}
}Built from the hub record. Check it against the publisher before submitting.
Added to the ChatPPG research hub
