This study proposes PEG-Prompt, a Paul-Elder critical thinking-guided prompt framework for LLM-based automated course project report (CPR) assessment. It evaluates CPRs across six dimensions. Experiments on the PEG-CPR dataset show PEG-Prompt significantly improves score alignment with human raters, especially when combined with key content extraction and few-shot examples.

This pilot study explores the effects of NAO robot-assisted classrooms on autistic students’ classroom performance. Results show NAO significantly improves online attention, communication, interaction assessment, and emotional states. Most students exhibit better engagement and stability, supporting NAO as an effective tool for autism spectrum disorder (ASD) group classroom intervention.

Structures of the human M-channel reveal a remarkably flexible assembly and the molecular basis for its subthreshold activation, enabling the design of a highly selective activator for precise epilepsy therapy.

A new study in Science Bulletin presents DVSTP, a deep learning system that integrates pathology images with spatial transcriptomics and proteomics to map intra-tumor heterogeneity. DVSTP predicts molecular profiles from routine pathology slides, making spatial multi-omics more accessible. Whole–tumor 3D reconstruction reveals that SRSF6 drives immune exclusion and is associated with poor clinical outcomes.

Large language models and autonomous agents have advanced rapidly, showing broad promise in medical imaging analysis, clinical diagnosis, and treatment planning. However, most existing medical AI systems still rely primarily on pre-trained knowledge and fixed workflows, making it difficult to learn continuously from long-term clinical feedback, patient outcomes, and prior treatment experience. This "static AI" architecture limits their value in complex real-world clinical settings.

To address this bottleneck, a team led by Dr. Lian Zhang from the First Hospital of Hebei Medical University, in collaboration with domestic and international research partners, has proposed VIBEMed, which is a self-evolving multi-agent framework for clinical decision support designed to enable dynamic learning and safe, traceable system evolution.

Within the neurovascular unit, neurons, glial cells, brain microvascular endothelial cells, and pericytes all harbor mitochondria. When these organelles become dysfunctional, they drive and amplify neuroinflammation in various ways. A new review in Science Bulletin provides a systematic framework for summarizing nanomedicines that can precisely target damaged mitochondria, revealing a promising strategy for the treatment of central nervous system diseases.