Tag
HiPPT是一个生理学基础的可穿戴PPG表示学习框架，通过多范围文本对齐（记录级形态对齐与跨记录上下文对齐）在超过400万条腕部PPG记录上预训练，在血压、血管僵硬度等六项下游任务的线性探测中取得最佳结果，并支持年龄组、性别和BMI的零样本画像。
REFIT introduces a label-free input calibration method for frozen activity-recognition models that corrects wearable sensor placement shifts (e.g., switching wrists or reattaching straps) by fitting axis-transform families to match training statistics, outperforming existing test-time adaptation approaches without retraining.
LiteEMG-FM is an efficient hybrid CNN-Transformer foundation model pretrained on 16 EMG datasets for robust, zero-calibration gesture recognition. The paper demonstrates real-time deployability on resource-constrained edge devices via a hierarchical wake-up gate and evaluates latency, power, and memory trade-offs across offloading paradigms.
This paper evaluates Jev, a fixed general-purpose probabilistic decision model, for training-free human activity recognition on accelerometer data, finding it far below supervised models (macro-F1 0.038–0.118 vs 0.686–0.907) and showing that adding more numerical sensor features actually degrades performance unless paired with deterministic semantic renderings.
A comparative study of temporal deep learning architectures for physiological emotion recognition using multimodal wearable datasets, evaluating LSTM, TCN, and Transformer models under different sensing configurations.
This paper proposes a coverage-aware virtual IMU augmentation framework to address data scarcity in human activity recognition, generating and selecting synthetic sensor data to improve model performance with limited labeled data.
This paper introduces the Individual Conformal Coupling Monitor (ICCM), a pre-inference method to detect structural ambiguity in wearable stress classification, improving safety by routing uncertain signals for abstention or deferral.
This paper presents a longitudinal multimodal study of 66 older adults using wearable sensing and clinical assessments to predict physical activity, sleep duration, and sleep apnea severity, finding that behavioral targets are more predictable and historical features are key predictors.