Search papers, labs, and topics across Lattice.
This study introduces a context-aware learning framework that enhances defect classification in atomic-resolution scanning transmission electron microscopy (STEM) by integrating image contrast with contextual metadata such as composition and imaging conditions. By leveraging a dataset of approximately 55 million simulated patches, the authors demonstrate that incorporating contextual variables significantly improves classification accuracy, achieving over 98% accuracy on simulations and near-human agreement on experimental data. The approach not only reduces posterior entropy by 94% but also establishes a more physically grounded method for defect classification, paving the way for advanced multimodal AI applications in materials characterization.
Contextual grounding in defect classification can elevate accuracy to over 98%, transforming a traditionally ambiguous task into a precise science.
Artificial intelligence is rapidly advancing materials characterization, yet most applications in electron microscopy rely solely on image contrast, overlooking the chemical and experimental context that shapes image formation. This limitation makes defect classification inherently ambiguous, as similar contrasts can arise from different materials or imaging conditions. Here we develop a context-aware learning framework that integrates image-derived contrast with metadata describing composition, beam energy, and detector geometry. Using a systematically constructed dataset of ~55 million simulated patches spanning 576 cases across 96 doped monolayer transition-metal dichalcogenides, we show that conditioning on contextual variables transforms defect classification from an ill-posed image-only task into a well-posed, physically grounded problem. The framework achieves over 98% accuracy on simulations and near-human agreement on experimental data, with a 94% reduction in posterior entropy. By emphasizing contextual grounding over architectural complexity, this approach links experimental image contrast to the underlying chemical and imaging conditions, supporting physically grounded defect assignments and a general pathway toward multimodal AI models for autonomous materials characterization.
NUS