# PULSE: Privileged Knowledge Transfer from Rich to Deployable Sensors for Embodied Multi-Sensory Learning

arXiv: https://arxiv.org/abs/2510.24058 · HTML: https://mlsanity.com/paper/2510.24058/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 35 references that are in the map.
- First posted on arXiv 28 Oct 2025.

## Builds on (most cited first, 17 of 17)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Time-Series Representation Learning via Temporal and Contextual Contrasting](https://mlsanity.com/paper/2106.14112/) (2021)
- [Self-Supervised Contrastive Pre-Training For Time Series via Time-Frequency Consistency](https://mlsanity.com/paper/2206.08496/) (2022)
- [Unifying distillation and privileged information](https://mlsanity.com/paper/1511.03643/) (2015)
- [CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients](https://mlsanity.com/paper/2005.13249/) (2020)
- [A Transformer Architecture for Stress Detection from ECG](https://mlsanity.com/paper/2108.09737/) (2021)
- [Frequency-Aware Masked Autoencoders for Multimodal Pretraining on Biosignals](https://mlsanity.com/paper/2309.05927/) (2023)
- [Wearable Accelerometer Foundation Models for Health via Knowledge Distillation](https://mlsanity.com/paper/2412.11276/) (2024)
- [Distilling Privileged Multimodal Information for Expression Recognition using Optimal Transport](https://mlsanity.com/paper/2401.15489/) (2024)
- [From the Lab to the Wild: Affect Modeling via Privileged Information](https://mlsanity.com/paper/2305.10919/) (2023)
- [Privileged Information for Modeling Affect In The Wild](https://mlsanity.com/paper/2107.10552/) (2021)
- [On the Generalizability of ECG-based Stress Detection Models](https://mlsanity.com/paper/2210.06225/) (2022)
- [ADAPT: Multimodal Learning for Detecting Physiological Changes under Missing Modalities](https://mlsanity.com/paper/2407.03836/) (2024)
- [Multi Teacher Privileged Knowledge Distillation for Multimodal Expression Recognition](https://mlsanity.com/paper/2408.09035/) (2024)
- [Modally Reduced Representation Learning of Multi-Lead ECG Signals through Simultaneous Alignment and Reconstruction](https://mlsanity.com/paper/2405.19359/) (2024)
- [Cross-Modality Investigation on WESAD Stress Classification](https://mlsanity.com/paper/2502.18733/) (2025)

## Abstract (arXiv)

Multi-sensory systems for embodied intelligence, from wearable body-sensor networks to instrumented robotic platforms, routinely face a sensor-asymmetry problem: the richest modality available during laboratory data collection is absent or impractical at deployment time due to cost, fragility, or interference with physical interaction. We introduce PULSE, a general framework for privileged knowledge transfer from an information-rich teacher sensor to a set of cheaper, deployment-ready student sensors. Each student encoder produces shared (modality-invariant) and private (modality-specific) embeddings; the shared subspace is aligned across modalities and then matched to representations of a frozen teacher via multi-layer hidden-state and pooled-embedding distillation. Private embeddings preserve modality-specific structure needed for self-supervised reconstruction, which we show is critical to prevent representational collapse. We instantiate PULSE on the wearable stress-monitoring task, using electrodermal activity (EDA) as the privileged teacher and ECG, BVP, accelerometry, and temperature as students. On the WESAD benchmark under leave-one-subject-out evaluation, PULSE achieves 0.994 AUROC and 0.988 AUPRC (0.965/0.955 on STRESS) without EDA at inference, exceeding all no-EDA baselines and matching the performance of a full-sensor model that retains EDA at test time. We further demonstrate modality-agnostic transfer with ECG as teacher, provide extensive ablations on hidden-state matching depth, shared-private capacity, hinge-loss margin, fusion strategy, and modality dropout, and discuss how the framework generalizes to broader embodied sensing scenarios involving tactile, inertial, and bioelectrical modalities.