ChemFixer: Correcting Invalid Molecules to Unlock Previously Unseen Chemical Space

Deep learning is significantly impacting microscopy, but traditional models often struggle with images from new instruments or settings. The study introduces a novel approach called Subnetwork Image Translation ADDA with automatic depth selection (SIT-ADDA-Auto), which adapts only the initial convolutional layers while keeping deeper layers unchanged. This method enhances transfer reliability and integrates predictive uncertainty to determine adaptation depth without needing target labels. The framework demonstrates improved reconstruction across various conditions, showcasing its robustness t…

The paper introduces a novel likelihood-guided variational Ising-based regularization framework for Vision Transformers (ViTs), aimed at enhancing model generalization while dynamically pruning redundant parameters. This approach utilizes Bayesian sparsification techniques to impose structured sparsity on model weights, allowing for adaptive architecture search during training. Unlike traditional dropout methods, this framework learns task-adaptive regularization, improving efficiency and interpretability in classification tasks involving structured and high-dimensional data.

Unitho is a unified multi-task framework designed for computational lithography, leveraging the Transformer architecture. It addresses critical tasks such as mask generation, rule violation detection, and layout optimization, which have traditionally been performed in isolation due to limited datasets. Trained on a large-scale industrial lithography simulation dataset comprising hundreds of thousands of cases, Unitho demonstrates significant effectiveness and generalizability, outperforming academic baselines in experimental results.