Over the past two decades, Microsoft Research India has achieved an extraordinary record of innovation—in areas ranging from health and education to agriculture and accessibility.
Yadong Lu, Senior Researcher; Thomas Dhome-Casanova (opens in new tab), Software Engineer; Jianwei Yang, Principal Researcher; Ahmed Awadallah, Partner Research Manager Graphic User interface (GUI) automation requires agents with the ability to understand and interact with user screens. However, using…
PIKE-RAG：解锁领域私有数据价值，推动LLMs工业应用落地
编者按：随着人工智能技术的快速发展，检索增强生成（RAG）系统已成为扩展大语言模型（LLMs）能力的重要手段之一。然而，当这些系统应用于复杂多样的工业场景时，仍面临诸多挑战，尤其是在处理领域特定知识和复杂推理任务时。对此，微软亚洲研究院的研究员们提出了 PIKE-RAG，通过多层次异构知识库构建、任务驱动的系统搭建策略以及自我进化的领域知识学习机制，显著提升了 LLMs 在复杂工业场景中的推理和应...
ExACT: Improving AI agents’ decision-making via test-time compute scaling
| Baolin Peng, Xiao Yu, Hao Cheng, Michel Galley, Zhou Yu, and Jianfeng Gao
ExACT combines Reflective-MCTS and Exploratory Learning to improve AI agents' decision-making, enabling test-time compute scaling. Learn how these methods help agents refine strategies for state-of-the-art performance and improved computational efficiency.
In the news | LinkedIn News
HeritageWatch.ai
Microsoft and our partners are launching HeritageWatch.AI, a nonprofit that will use AI to protect cultural heritage sites across the globe and help us preserve humanity’s past for future generations.
Ideas: Building AI for population-scale systems with Akshay Nambi
| Chris Stetkiewicz and Akshay Nambi
Advances in AI are driving meaningful real-world impact. Principal Researcher Akshay Nambi shares how his passion for tackling real-world challenges across various domains fuels his work in building reliable and robust AI systems.
In the news | Harvard Business School
America’s Digital Divide: Where Workers Are Falling Behind
With rural areas significantly lagging behind cities in computer use, research by Raffaella Sadun, Shane Greenstein, and colleagues finds that many Americans lack the digital literacy that's increasingly needed in an AI world.
MobiCom is one of the premier international academic conferences in the field of mobile computing and wireless networks. In this article, we select several papers from Microsoft Research Asia that were accepted at MobiCom 2024. These papers explore a diverse range…
Advances to low-bit quantization enable LLMs on edge devices
| Shijie Cao, Lingxiao Ma, and Ting Cao
Advances in low-bit quantization techniques enable efficient operation of LLMs on resource-constrained edge devices. Discover how innovations like T-MAC, Ladder, and LUT Tensor Core improve computational efficiency and enhance hardware compatibility.