# PI-TTA: Physics-Informed Source-Free Test-Time Adaptation for Robust Human Activity Recognition on Mobile Devices

arXiv: https://arxiv.org/abs/2604.25435 · HTML: https://mlsanity.com/paper/2604.25435/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 46 references that are in the map.
- First posted on arXiv 28 Apr 2026.

## Builds on (most cited first, 9 of 9)

- [Hamiltonian Neural Networks](https://mlsanity.com/paper/1906.01563/) (2019)
- [Revisiting Batch Normalization For Practical Domain Adaptation](https://mlsanity.com/paper/1603.04779/) (2016)
- [OpenFedLLM: Training Large Language Models on Decentralized Private Data via Federated Learning](https://mlsanity.com/paper/2402.06954/) (2024)
- [Tent: Fully Test-time Adaptation by Entropy Minimization](https://mlsanity.com/paper/2006.10726/) (2020)
- [Temporal Test-Time Adaptation with State-Space Models](https://mlsanity.com/paper/2407.12492/) (2024)
- [Diversity-aware Buffer for Coping with Temporally Correlated Data Streams in Online Test-time Adaptation](https://mlsanity.com/paper/2401.00989/) (2024)
- [When and Where to Reset Matters for Long-Term Test-Time Adaptation](https://mlsanity.com/paper/2603.03796/) (2026)
- [Order-Aware Test-Time Adaptation: Leveraging Temporal Dynamics for Robust Streaming Inference](https://mlsanity.com/paper/2601.21012/) (2026)
- [DATTA: Domain Diversity Aware Test-Time Adaptation for Dynamic Domain Shift Data Streams](https://mlsanity.com/paper/2408.08056/) (2024)

## Built on by (most cited first, 2 of 2)

- [When Adaptation Hurts: Split Sensitivity and Person-Level Negative Transfer in Federated Wearable Onboarding](https://mlsanity.com/paper/2609.27819/) (2026)
- [REFIT: Recognize, Fix, and Test Wearable Sensor Placement Shifts without Labels](https://mlsanity.com/paper/2610.08991/) (2026)

## Abstract (arXiv)

Source-free test-time adaptation (TTA) is appealing for mobile and wearable sensing because it enables on-device personalization from unlabeled test streams without centralizing private data. However, sensor-based human activity recognition (HAR) poses challenges that are less pronounced in standard vision benchmarks: behavioral inertial streams are temporally correlated and often exhibit within-session shifts caused by sensor rotation, placement change, and sampling-rate drift. Under this streaming non-i.i.d. setting, widely used vision-style TTA objectives can become unstable, leading to overconfident errors, representation collapse, and catastrophic forgetting. We propose PI-TTA, a lightweight source-free adaptation framework that stabilizes online updates through three physics-consistent constraints: gravity consistency, short-horizon temporal continuity, and spectral stability. PI-TTA updates the same small parameter subset as strong source-free baselines and incurs only modest overhead, making it suitable for on-device deployment. Experiments on USCHAD, PAMAP2, and mHealth under long-sequence stress tests and factorized shift protocols show that PI-TTA mitigates the severe degradation observed in confidence-driven baselines and preserves stable adaptation under sustained streaming conditions. It improves long-sequence accuracy by up to 9.13% and reduces physical-violation rates by 27.5%, 24.1%, and 45.4% on USCHAD, PAMAP2, and mHealth, respectively. These results demonstrate that physics-informed adaptation can improve accuracy, stability, and deployment reliability for real-world mobile sensing systems.