Answers are grounded in this paper and cite their sources.
Learning-based approach improves diagnostic accuracy and latency in AR/VR and wearable computing systems, suggesting a new method for fault analysis.
AR/VR and wearable computing systems have become increasingly complex, integrating diverse sensors, hardware modules, and distributed computing pipelines. As system scale grows, diagnosing root causes of anomalies becomes difficult due to high heterogeneity, multimodal data sources, and implicit inter-module dependencies. This paper presents a learning-based root cause prediction framework that leverages multimodal error logs, performance metrics, and sensor streams to automatically identify and interpret system failures. The framework integrates log parsing, temporal embedding, reliability-aware multimodal fusion, and causal graph propagation to provide interpretable predictions. Experiments on simulated and real-world AR/VR devices demonstrate substantial improvements in diagnostic accuracy and latency compared to traditional single-modality or rule-based methods. Our approach advances the frontier of autonomous fault analysis in resource-constrained intelligent systems.
No takes yet. Share an insight, caveat, or question.
Changsheng Wu (2025) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: