Toward World Modeling of Physiological Signals with Chaos-Theoretic Balancing and Latent Dynamics

2026-05-14

AI Summary

Existing approaches to physiological signal modeling are often confined to static tasks and struggle to capture long-term dynamics and intervention effects. This work proposes NormWear-2, a world model that maps multivariate physiological signals and clinical interventions into a shared latent space, integrating pretrained priors with nonparametric state adjustments to enable constrained, coherent predictions across multiple timescales. The method innovatively incorporates a chaos-theory-guided data balancing mechanism to enhance representation robustness and establishes a novel dynamic systems modeling paradigm that unifies intuition with analytical insight. Evaluated on a multi-scenario dataset encompassing 8,026 subjects, NormWear-2 significantly outperforms state-of-the-art temporal foundation models in time-domain, frequency-domain, and latent-space metrics while maintaining strong representational capacity for downstream tasks.

Abstract

Physiological time series signals reflect complex, multi-scale dynamical processes of the human body. Existing modeling studies focus on static tasks such as classification, event forecasting, or short-horizon next step prediction, while long-horizon signal-level forecasting and predictive nature of physiological signals remain underexplored. We introduce NormWear-2, a world model that encodes both multivariate physiological signals and clinical intervention variables into a shared latent space and models their joint temporal evolution as a dynamical system. Our approach combines inference from prior pre-trained knowledge (intuition) with instant non-parametric latent state transition adaptation (insight), enabling coherent forecasting across multiple temporal scales, conditioned on heterogeneous clinical interventions. During the pretraining phase, we find that chaos-theoretic balancing of dynamical regime diversity yields more robust representations, with a smaller balanced corpus outperforming one twice its size and capturing bifurcation regimes. We evaluate the world model performance across diverse real-world physiological datasets spanning heterogeneous temporal resolutions and intervention regimes, covering daily life, point-of-care, and clinical settings, including fitness planning, hemodialysis, diabetes management, and surgical monitoring. These evaluation datasets comprise records from 8,026 subjects, spanning study durations from 3.2 hours for high-resolution signal data to 2.3 years for longitudinal clinical biomarker tracking. NormWear-2 achieves the best overall forecasting performance across time, frequency, and latent representation domains, with significant improvements over state-of-the-art time series foundation models, while maintaining competitive downstream representation quality, providing a step toward general-purpose world models for physiological signals.

Problem

• physiological signals
• long-horizon forecasting
• world modeling
• dynamical systems
• time series prediction

Innovation

• world model
• chaos-theoretic balancing
• latent dynamics
• physiological time series
• non-parametric adaptation