Clinician skepticism toward opaque AI hinders adoption in high-stakes healthcare. We present AICare, an interactive and interpretable AI copilot for collaborative clinical decision-making. By analyzing longitudinal electronic health records, AICare grounds dynamic risk predictions in scrutable visualizations and LLM-driven diagnostic recommendations. Through a within-subjects counterbalanced study with 16 clinicians across nephrology and obstetrics, we comprehensively evaluated AICare using objective measures (task completion time and error rate), subjective assessments (NASA-TLX, SUS, and confidence ratings), and semi-structured interviews. Our findings indicate AICare’s reduced cognitive workload. Beyond performance metrics, qualitative analysis reveals that trust is actively constructed through verification, with interaction strategies diverging by expertise: junior clinicians used the system as cognitive scaffolding to structure their analysis, while experts engaged in adversarial verification to challenge the AI’s logic. This work offers design implications for creating AI systems that function as transparent partners, accommodating diverse reasoning styles to augment rather than replace clinical judgment.

Virtual reality (VR) allows users to observe and manipulate 3D geometry from multiple viewpoints. Most VR selection work, however, optimizes techniques for selecting entire objects. Selecting a single face on a polyhedron remains underexplored and is more challenging because the interaction must act on a local component while preserving the object's global structure. We introduce a design space tailored to this task with three dimensions: viewing strategy, disambiguation consistency, and interaction metaphor. Guided by this space, we design eight freehand techniques for polyhedral face selection. A within-subjects study with 16 participants evaluates these techniques across polyhedral complexity (two radii; face counts 4, 6, and 12). The results identify three complementary top techniques, reveal tradeoffs between viewing choices and geometric preservation, and yield concrete guidelines for matching techniques to target geometry and task demands. A follow-up study with complex, realistic models confirms the robustness and practical usability of the three techniques. Together, these contributions shift attention from whole object selection to precise component selection in VR and provide actionable methods for 3D modeling, assembly, and texturing.

Foreground-conditioned inpainting aims to seamlessly fill the background region of an image by utilizing the provided foreground subject and a text description. While existing T2I-based image inpainting methods can be applied to this task, they suffer from issues of subject shape expansion, distortion, or impaired ability to align with the text description, resulting in inconsistencies between the visual elements and the text description. To address these challenges, we propose Pinco, a plug-and-play foreground-conditioned inpainting adapter that generates high-quality backgrounds with good text alignment while effectively preserving the shape of the foreground subject. Firstly, we design a Self-Consistent Adapter that integrates the foreground subject features into the layout-related self-attention layer, which helps to alleviate conflicts between the text and subject features by ensuring that the model can effectively consider the foreground subject's characteristics while processing the overall image layout. Secondly, we design a Decoupled Image Feature Extraction method that employs distinct architectures to extract semantic and spatial features separately, significantly improving subject feature extraction and ensuring high-quality preservation of the subject's shape. Thirdly, to ensure precise utilization of the extracted features and to focus attention on the subject region, we introduce a Shared Positional Embedding Anchor, greatly improving the model's understanding of subject features and boosting training efficiency. Extensive experiments demonstrate that our method achieves superior performance and efficiency in foreground-conditioned inpainting.

Medical entity normalization is an important task for medical information processing. The Unified Medical Language System (UMLS), a well-developed medical terminology system, is crucial for medical entity normalization. However, the UMLS primarily consists of English medical terms. For languages other than English, such as Chinese, a significant challenge for normalizing medical entities is the lack of robust terminology systems. To address this issue, we propose a translation-enhancing training strategy that incorporates the translation and synonym knowledge of the UMLS into a language model using the contrastive learning approach. In this work, we proposed a cross-lingual pre-trained language model called TeaBERT, which can align synonymous Chinese and English medical entities across languages at the concept level. As the evaluation results showed, the TeaBERT language model outperformed previous cross-lingual language models with Acc@5 values of 92.54%, 87.14% and 84.77% on the ICD10-CN, CHPO and RealWorld-v2 datasets, respectively. It also achieved a new state-of-the-art cross-lingual entity mapping performance without fine-tuning. The translation-enhancing strategy is applicable to other languages that face the similar challenge due to the absence of well-developed medical terminology systems.

Recently emerged heterophilous graph neural networks have significantly reduced the reliance on the assumption of graph homophily where linked nodes have similar features and labels. These methods focus on a supervised setting that relies on labeling information heavily and presents the limitations on general graph downstream tasks. In this work, we propose a self-supervised representation learning paradigm on graphs with heterophily (namely HGRL) for improving the generalizability of node representations, where node representations are optimized without any label guidance. Inspired by the designs of existing heterophilous graph neural networks, HGRL learns the node representations by preserving the node original features and capturing informative distant neighbors. Such two properties are obtained through carefully designed pretext tasks that are optimized based on estimated high-order mutual information. Theoretical analysis interprets the connections between HGRL and existing advanced graph neural network designs. Extensive experiments on different downstream tasks demonstrate the effectiveness of the proposed framework.

Estimating the accuracy of an automatically constructed knowledge graph (KG) becomes a challenging task as the KG often contains a large number of entities and triples. Generally, two major components information extraction (IE) and entity linking (EL) are involved in KG construction. However, the existing approaches just focus on evaluating the triple accuracy that indicates the IE quality, completely ignoring the entity accuracy. Motivated by the fact that the major advance of machines is the strong computing power while humans are skilled in correctness verification, we propose an efficient interactive method to reduce the overall cost for evaluating the KG quality, which produces accuracy estimates with a statistical guarantee for both triples and entities. Instead of annotating triples and entities separately, we design a general annotation cost that blends triples and entities generated from the identical source text. During human verification, the machine can pre-compute and infer triples to be annotated in the next round by speculating human feedback. The human-machine collaborative mechanism is optimized by formulating an order selection problem of triples which is NP-hard. Thus, a Monte Carlo Tree Search is proposed to guide the annotation process by finding an approximate solution. Extensive experiments demonstrate that our method takes less annotation cost while yielding higher accuracy estimation quality compared to the state-of-the-art approaches.

As a hyper-natural interaction technique in 3D user interfaces, non-isomorphic rotation has been considered an effective approach for rotation tasks, where a static or dynamic control-display gain can be applied to amplify or attenuate a rotation. However, it is not clear whether non-isomorphic rotation can benefit 6-degree-of-freedom (6-DOF) manipulation tasks in AR and VR. In this article, we extended the usability studies of non-isomorphic rotation from rotation-only tasks to 6-DOF manipulation tasks and analyzed the collected data using a 2-component model. Using a mixed reality (MR) simulation approach, we also investigated whether environment (AR or VR) had an impact on 3D manipulation tasks. The results reveal that although both static and dynamic non-isomorphic rotation techniques could save time and effort in ballistic phases, only dynamic non-isomorphic rotation was significantly faster than isomorphic rotation. Interestingly, while environment had no significant impact on overall user performance, we found evidence that it could affect fine-tuning in correction phases. We also found that most participants preferred AR over VR, indicating that environmental visual realism could be helpful to improve user experience.

The industry and academia have proposed many distributed graph processing systems. However, the existing systems are not friendly enough for users like data analysts and algorithm engineers. On the one hand, the programming models and interfaces differ a lot in the existing systems, leading to high learning costs and program migration costs. On the other hand, these graph processing systems are tightly bound to the underlying distributed computing platforms, requiring users to be familiar with distributed computing. To improve the usability of distributed graph processing, we propose a unified distributed graph programming framework UniGPS. Firstly, we propose a unified cross-platform graph programming model VCProg for UniGPS. VCProg hides details of distributed computing from users. It is compatible with the popular graph programming models Pregel, GAS, and Push-Pull. VCProg-based programs can be executed by compatible distributed graph processing systems without modification, reducing the learning overheads of users. Secondly, UniGPS supports Python as the programming language. We propose an interprocess-communication-based execution environment isolation mechanism to enable Java/C++-based graph processing systems to call user-defined methods written in Python. The experimental results show that UniGPS enables users to process big graphs beyond the memory capacity of a single machine without sacrificing usability. UniGPS shows near-linear data scalability and machine scalability.