Limited-angle computed tomography (LACT) improves temporal resolution and reduces radiation dose, but suffers from severe artifacts due to missing projections. Clinical workflows record abundant patient- and acquisition-level metadata, yet such information remains underutilized in image reconstruction. To tackle the ill-posed LACT inverse problem, we propose a metadata-guided two-stage diffusion framework that leverages structured clinical contexts as semantic priors for robust reconstruction. In Stage-I, we learn a metadata-to-anatomy generative prior by conditioning a transformer-based diffusion model on clinical metadata (acquisition parameters, patient demographics, and diagnostic impressions), and sampling a coarse anatomical estimate from Gaussian noise. In Stage-II, a second conditional diffusion model performs coarse-to-fine refinement, using the Stage-I estimate as an image prior while re-injecting the same metadata to recover full-resolution anatomy. To preserve anatomical fidelity and suppress hallucinations, projection-domain data consistency is enforced periodically after denoising update via an ADMM-based solver. Experiments on the public multimodal CTRATE dataset demonstrate that the proposed framework outperforms iterative, CNN-based, and diffusion-based baselines, with the greatest gains under severe truncation, e.g., up to 5.23%/11.21% higher SSIM/PSNR than the strongest metadata-free diffusion competitor at 90°. On real clinical cardiac CT, it yields coronary artery calcium scores closer to full-view references, indicating improved clinical utility. Furthermore, the proposed method is generalized to out-of-distribution angular ranges and projection geometries, and ablation results confirm complementary contributions from different metadata types under limited-angle conditions. Our results highlight clinical metadata as actionable semantic priors to synergize with physics-informed constraints to improve both reconstruction fidelity and clinical quantification in LACT.

Current graph representation (GR) algorithms require huge demand of human experts in hyperparameter tuning, which significantly limits their practical applications, leading to an urge for automated graph representation without human intervention. Although automated machine learning (AutoML) serves as a good candidate for automatic hyperparameter tuning, little literature has been reported on automated graph presentation learning and the only existing work employs a black-box strategy, lacking insights into explaining the relative importance of different hyperparameters. To address this issue, we study explainable automated graph representation with hyperparameter importance in this paper. We propose an explainable AutoML approach for graph representation (e-AutoGR) which utilizes explainable graph features during performance estimation and learns decorrelated importance weights for different hyperparameters in affecting the model performance through a non-linear decorrelated weighting regression. These learned importance weights can in turn help to provide more insights in hyperparameter search procedure. We theoretically prove the soundness of the decorrelated weighting algorithm. Extensive experiments on real-world datasets demonstrate the superiority of our proposed e-AutoGR model against state-of-the-art methods in terms of both model performance and hyperparameter importance explainability.

Split keyboards are widely used on hand-held touchscreen devices (e.g., tablets). However, typing on a split keyboard often requires eye movement and attention switching between two halves of the keyboard, which slows users down and increases fatigue. We explore peripheral typing, a superior typing mode in which a user focuses her visual attention on the output text and keeps the split keyboard in peripheral vision. Our investigation showed that peripheral typing reduced attention switching, enhanced user experience and increased overall performance (27 WPM, 28% faster) over the typical eyes-on typing mode. This typing mode can be well supported by accounting the typing behavior in statistical decoding. Based on our study results, we have designed GlanceType, a text entry system that supported both peripheral and eyes-on typing modes for real typing scenario. Our evaluation showed that peripheral typing not only well co-existed with the existing eyes-on typing, but also substantially improved the text entry performance. Overall, peripheral typing is a promising typing mode and supporting it would significantly improve the text entry performance on a split keyboard.

Modern touchscreen keyboards are all powered by the word-level auto-correction ability to handle input errors. Unfortunately, visually impaired users are deprived of such benefit because a screen-reader keyboard offers only character-level input and provides no correction ability. In this paper, we present VIPBoard, a smart keyboard for visually impaired people, which aims at improving the underlying keyboard algorithm without altering the current input interaction. Upon each tap, VIPBoard predicts the probability of each key considering both touch location and language model, and reads the most likely key, which saves the calibration time when the touchdown point misses the target key. Meanwhile, the keyboard layout automatically scales according to users' touch point location, which enables them to select other keys easily. A user study shows that compared with the current keyboard technique, VIPBoard can reduce touch error rate by 63.0% and increase text entry speed by 12.6%.