Research
Empowering everyone to understand digital content provenance through human-centric design “Generative AI is blurring the boundary between authentic and synthetic media. Without accessible, human-centered provenance tools, we risk drifting into a post-epistemic world where fact…
Microsoft Research Asia Honorary Scholars 辰星荣誉计划（Honorary Scholars Talent Program）是微软亚洲研究院（Microsoft Research Asia, MSRA）面向顶尖科研学子推出的高潜人才专项，我们希望在全球范围内寻找和我们一起定义AI新纪元的同行人。 本计划首次开…
Cryptographically securing the authenticity of digital assets As generative AI accelerates the creation of realistic synthetic media (e.g., photos, videos, audio, documents), it is increasingly difficult for people and platforms to distinguish authentic content from…
MSR Synth is a research project exploring how to design AI agents with social intelligence—agents that can communicate, collaborate, and participate in human interaction in ways that are context‑aware, trustworthy, and aligned with human values.…
This project studies the data protection challenges when models and agents interact with sensitive data across multiple users. Large language models (LLMs) are increasingly deployed in enterprise settings where thLarge language models (LLMs) are increasingly…
Communication-centered computing Communication shapes how societies create, share, and preserve knowledge. Yet today’s digital tools remain organized around static formats that enforce rigid separations between creation and consumption, author and reader, expression and interpretation. These…
< AI for Good Lab Language is more than communication—it sustains culture, identity, and opportunity. Yet most of the world’s languages remain underrepresented in today’s AI systems because they have limited digital content, few high-quality…
Automated evaluation and testing of AI agents AI agents are rapidly transforming software, with projections of over a billion agents in operation by 2028. These agents, embedded in products like VS Code and M365 Copilot,…
Small foundation models for the electric grid GridFM is a Microsoft Research initiative to build a foundation model (FM) for electric power grids, applying modern AI methods—similar to large language/weather models—to complex grid physics. Traditional…
This project aims to push the boundaries of agentic AI by addressing three critical challenges: long-horizon memory, safe and aligned tool usage, and adaptive reasoning. Current language models excel at text generation but lack agency—the…