Coauthor roundtable: Reflecting on healthcare economics, biomedical research, and medical education
| Peter Lee, Carey Goldberg, and Dr. Isaac Kohane
For the series finale, Peter Lee, Carey Goldberg, and Dr. Zak Kohane compare their predictions to insights from the series’ most recent guests, including experts on AI’s economic and societal impact, leaders in AI-driven medicine, and doctors in training.
编者按：欢迎阅读“开源上新”栏目！“开源上新”聚焦展示微软亚洲研究院最新发布的开源项目以及开源项目中的重大功能更新。作为“科研上新”的姊妹篇，“开源上新”栏目将帮助你快速了解开源项目的核心价值、关键功能与应用场景，让你能够掌握前沿技术、获取实践案例、提升开发效率，更好地将这些开源项目应用到自己的工作与研究中。 本期我们带来的是微软亚洲研究院全新开源的 Agent Lightning。它开创性地提出...
MindJourney enables AI to explore simulated 3D worlds to improve spatial interpretation
| Yuncong Yang, Reuben Tan, Swadheen Shukla, and Jianfeng Gao
MindJourney can enable AI to navigate and interpret 3D environments from limited visual input, potentially improving performance in navigation, planning, and safety-critical tasks.
In the news | Fierce Healthcare
Epic unveils AI solutions for clinicians, patients and RCM, gives sneak peek at Cosmos AI for risk prediction
Imagine a pair of smart glasses that detects its surroundings and speaks up at critical moments, such as when a car is approaching. That kind of split-second assistance could be transformative for people with low vision, but today’s visual AI…
HeurAgenix：大语言模型驱动的组合优化“全能教练”
编者按：在物流规划、生产排程、能源调度等行业应用场景中，组合优化问题往往规模庞大、约束复杂，长期依赖专家手工调参的启发式算法已难以应对动态多变的现实需求。微软亚洲研究院最新提出的 HeurAgenix 框架，将大语言模型视作“全能教练”，为启发式算法带来了自进化与自适应的能力，并通过蒸馏压缩实现快速响应。HeurAgenix 不仅全面超越了现有 LLMs 超启发式方法，在多数任务中甚至优于专业求解...
CausCell：虚拟细胞进入“白盒时代”
编者按：在生命科学研究中，逼真且可解释的细胞模型一直备受期待。但现有的单细胞 AI 建模多依赖于深度神经网络等“黑盒”方法，难以解开高维组学数据背后的生物学因果机制。近日，微软亚洲研究院（上海）与同济大学合作在 《自然-通讯》（Nature Communications）杂志上发表了最新成果 CausCell。这一框架首次将结构因果模型与扩散模型深度融合，实现了细胞尺度的因果解耦表征与可控的反事实...
Dion: the distributed orthonormal update revolution is here
| Kwangjun Ahn and John Langford
Dion is a new AI model optimization method that boosts scalability and performance over existing leading methods by orthonormalizing only a top rank subset of singular vectors, enabling more efficient training of large models such as LLaMA-3 with reduced overhead.
Reimagining healthcare delivery and public health with AI
| Peter Lee, Dr. Umair Shah, and Dr. Gianrico Farrugia
Former Washington State Secretary of Health Dr. Umair Shah and Mayo Clinic CEO Dr. Gianrico Farrugia explore how healthcare leaders are approaching AI when it comes to public health, care delivery, the healthcare-research connection, and the patient experience.