# Temporal Structure Matters for Efficient Test-Time Adaptation in Wearable Human Activity Recognition

arXiv: https://arxiv.org/abs/2605.04617 · HTML: https://mlsanity.com/paper/2605.04617/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 36 references that are in the map.
- First posted on arXiv 6 May 2026.

## Builds on (most cited first, 12 of 12)

- [Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2002.08546/) (2020)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [Towards Stable Test-Time Adaptation in Dynamic Wild World](https://mlsanity.com/paper/2302.12400/) (2023)
- [A Comprehensive Survey on Test-Time Adaptation under Distribution Shifts](https://mlsanity.com/paper/2303.15361/) (2023)
- [Robust Test-Time Adaptation in Dynamic Scenarios](https://mlsanity.com/paper/2303.13899/) (2023)
- [Deep Learning for Sensor-based Human Activity Recognition: Overview, Challenges and Opportunities](https://mlsanity.com/paper/2001.07416/) (2020)
- [A Prototype-Oriented Framework for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2110.12024/) (2021)
- [Optimization-Free Test-Time Adaptation for Cross-Person Activity Recognition](https://mlsanity.com/paper/2310.18562/) (2023)
- [Online Human Activity Recognition Employing Hierarchical Hidden Markov Models](https://mlsanity.com/paper/1903.04820/) (2019)
- [Panda: Test-Time Adaptation with Negative Data Augmentation](https://mlsanity.com/paper/2511.10481/) (2025)
- [Order-Aware Test-Time Adaptation: Leveraging Temporal Dynamics for Robust Streaming Inference](https://mlsanity.com/paper/2601.21012/) (2026)
- [ZOTTA: Test-Time Adaptation with Gradient-Free Zeroth-Order Optimization](https://mlsanity.com/paper/2603.14254/) (2026)

## 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)
- [What Egocentric Video Embeddings Encode and How to Combine Them with Inertial Sensors for Human Activity Recognition](https://mlsanity.com/paper/2610.12926/) (2026)

## Abstract (arXiv)

Wearable human activity recognition (WHAR) models often suffer from performance degradation under real-world cross-user distribution shifts. Test-time adaptation (TTA) mitigates this degradation by adapting models online using unlabeled test streams, yet existing methods largely inherit assumptions from vision tasks and underexploit the inherent inter-window temporal structure in WHAR streams. In this paper, we revisit such temporal structure as a feature-conditioned inference signal rather than merely an output-space smoothing prior. We derive the insight that temporal continuity and observation-induced feature deviations provide complementary cues for determining when to preserve or release temporal inertia and where to route prediction refinement during likely transitions. Building upon this insight, we propose SIGHT, a lightweight and backpropagation-free TTA framework for WHAR, enabling real-time edge deployment. SIGHT estimates predictive surprise by comparing the current feature with a prototype-based expected state, and then uses the resulting feature deviation to guide geometry-aware transition routing based on prototype alignment and stream-level marginal habit tracking. Evaluations on real-world datasets confirm that SIGHT outperforms existing TTA baselines while reducing computational and memory costs.