In the news | The Sequence
Making Small Models Great Achieve GPT-o1 Levels in Math Reasoning with Microsoft rStar-Math
rStar-Math is a novel approach that significantly boosts the mathematical reasoning capabilities of small language models (SLMs). This innovative system enables SLMs to achieve performance levels comparable to, and even exceeding, OpenAI’s o1, despite a significantly smaller model size. This…
WANet与WALoss：突破量子化学中的哈密顿量计算瓶颈
编者按：量子化学中的哈密顿量计算一直是制约药物筛选与新材料设计效率的瓶颈。近期，微软研究院科学智能中心、耶鲁大学及加州大学戴维斯分校联合研究提出的全新损失函数 WALoss 和深度学习框架 WANet，为这一难题提供了创新的解决方案。通过优化计算过程，WANet 显著提高了哈密顿量预测的精度与效率，尤其在大分子系统中展示了强大的扩展能力。搭配 WALoss，WANet 则能够准确预测分子能量，并有...
The rapid development of AI has steadily advanced the field of text-to-video (T2V) generation, offering a rich and convenient video content creation experience and unlocking new possibilities in entertainment, education, and multimedia communication. Traditional T2V methods, however, are limited due…
提示词优化、FP4量化训练、灵巧机器人抓取策略蒸馏、视频高效灵活建模
编者按：欢迎阅读“科研上新”栏目！“科研上新”汇聚了微软亚洲研究院最新的创新成果与科研动态。在这里，你可以快速浏览研究院的亮点资讯，保持对前沿领域的敏锐嗅觉，同时也能找到先进实用的开源工具。 本期，我们为大家精心挑选了4篇前沿研究论文，内容涵盖大语言模型的提示词优化、FP4 量化的高效训练方法、灵巧机器人抓取的简化策略以及视频内容高效建模的 Tokenizer，感兴趣的朋友快来看看吧！ 本期内容速...
Research advances are driving real-world impact faster than ever, with the latest advances in AI transforming the way we all live, work, and think. In this episode, discover how precision health, multimodal AI agents, and innovative chemical synthesis models are…
LLMs for safe low-level programming
Aseem Rastogi and Pantazis Deligiannis talk about two technical results from ICSE 2025 on using large language models (LLMs) for safe low-level programming. The results demonstrate LLMs inferring machine-checkable memory safety invariants in legacy C code and how LLMs assist…
AutoGen v0.4: Reimagining the foundation of agentic AI for scale, extensibility, and robustness
Gagan Bansal introduces a transformative update to the AutoGen framework that builds on user feedback and redefines modularity, stability, and flexibility to empower the next generation of agentic AI research and applications.
Belief state transformers
John Langford talks about a new transformer architecture that generates compact belief states for goal-conditioned planning, enhancing planning algorithms' efficiency and effectiveness.
Jianwei Yang introduces Magma, a new multimodal agentic foundation model designed for UI navigation in digital environments and robotics manipulation in physical settings.