Reproduce “MURFormer: Attention-Based Lightweight Multi-Scale Uncertainty-Guided Refinement Framework for PPG Denoising” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Haram Kwon, Chaeeun Hwang, Seayul Choi, Sungho Jo

MURFormer denoises wearable photoplethysmography (PPG) by reconstructing the signal, estimating patch-level uncertainty, and refining regions unreliable in a second pass. Its encoder combines convolution with attention over short and long temporal spans. On CIME, under max-absolute scaling, it achieved 10.72 dB SNR and 0.82 bpm heart-rate MAE, outperforming the diffusion baseline; refinement gains were modest, and uncertainty’s contribution remains unisolated.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.