# Retrieval-Augmented Personalization with Foundation Models for Wearable Stress Detection

arXiv: https://arxiv.org/abs/2606.24985 · HTML: https://mlsanity.com/paper/2606.24985/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 41 references that are in the map.
- First posted on arXiv 23 Jun 2026.

## Builds on (most cited first, 23 of 23)

- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [MOMENT: A Family of Open Time-series Foundation Models](https://mlsanity.com/paper/2402.03885/) (2024)
- [Personalization of Large Language Models: A Survey](https://mlsanity.com/paper/2411.00027/) (2024)
- [User-LLM: Efficient LLM Contextualization with User Embeddings](https://mlsanity.com/paper/2402.13598/) (2024)
- [Self-supervised Learning for Human Activity Recognition Using 700,000 Person-days of Wearable Data](https://mlsanity.com/paper/2206.02909/) (2022)
- [PaPaGei: Open Foundation Models for Optical Physiological Signals](https://mlsanity.com/paper/2410.20542/) (2024)
- [K-EmoCon, a multimodal sensor dataset for continuous emotion recognition in naturalistic conversations](https://mlsanity.com/paper/2005.04120/) (2020)
- [Personalized Machine Learning for Robot Perception of Affect and Engagement in Autism Therapy](https://mlsanity.com/paper/1802.01186/) (2018)
- [User Embedding Model for Personalized Language Prompting](https://mlsanity.com/paper/2401.04858/) (2024)
- [A Transformer Architecture for Stress Detection from ECG](https://mlsanity.com/paper/2108.09737/) (2021)
- [MERBench: A Unified Evaluation Benchmark for Multimodal Emotion Recognition](https://mlsanity.com/paper/2401.03429/) (2024)
- [A Survey on User Behavior Modeling in Recommender Systems](https://mlsanity.com/paper/2302.11087/) (2023)
- [A Survey on Personality-Aware Recommendation Systems](https://mlsanity.com/paper/2101.12153/) (2021)
- [Beyond Sensor Data: Foundation Models of Behavioral Data from Wearables Improve Health Predictions](https://mlsanity.com/paper/2507.00191/) (2025)
- [Transformer-based Self-supervised Multimodal Representation Learning for Wearable Emotion Recognition](https://mlsanity.com/paper/2303.17611/) (2023)
- [Personalized Adaptive Meta Learning for Cold-start User Preference Prediction](https://mlsanity.com/paper/2012.11842/) (2020)
- [Multi-level Fusion of Wav2vec 2.0 and BERT for Multimodal Emotion Recognition](https://mlsanity.com/paper/2207.04697/) (2022)
- [Affective Computing Has Changed: The Foundation Model Disruption](https://mlsanity.com/paper/2409.08907/) (2024)
- [Personalized Adaptation with Pre-trained Speech Encoders for Continuous Emotion Recognition](https://mlsanity.com/paper/2309.02418/) (2023)
- [Enrolment-based personalisation for improving individual-level fairness in speech emotion recognition](https://mlsanity.com/paper/2406.06665/) (2024)
- [Beyond Speech and More: Investigating the Emergent Ability of Speech Foundation Models for Classifying Physiological Time-Series Signals](https://mlsanity.com/paper/2410.12645/) (2024)
- [Assessing Foundation Models' Transferability to Physiological Signals in Precision Medicine](https://mlsanity.com/paper/2412.03427/) (2024)

## Built on by (most cited first, 1 of 1)

- [RACE: Residual-Aware Test-Time Adaptation for Neighbor-Rich Time-Series Foundation Model Forecasting](https://mlsanity.com/paper/2610.00405/) (2026)

## Abstract (arXiv)

Personalization in wearable-based stress detection remains challenging due to substantial inter-individual variability in physiological and behavioral responses. While traditional approaches rely on user-specific fine-tuning or costly self-supervised pre-training on large datasets, we propose a lightweight alternative based on retrieval-augmented personalization. Our method leverages frozen, out-of-domain foundation models to retrieve similar patterns from a target user's history and encode them into a compact personalized embedding that modulates representations extracted by a lightweight transformer network. We evaluate our approach on the WESAD stress detection dataset with N=15 users, comprising wrist-worn physiological (EDA, BVP, temperature) and activity (accelerometer) signals, and report gains of +3.92% in accuracy and +4.76% in macro F1-score over a non-personalized transformer baseline, approaching supervised fine-tuning performance without requiring any labeled user data. We further show that temporal retrieval, where only prior user samples are available, achieves performance close to full intra-user retrieval, demonstrating robustness to limited user history. Finally, we explore personalization in a cross-dataset retrieval setting, leveraging embeddings from the K-Emocon dataset to personalize representations for stress detection on the WESAD dataset.