AI agents often fail because their instructions, or skills, are manually modified with no guarantee of improvement. Learn how SkillOpt turns skill editing into a training process, making agent behavior more reliable without changing model weights: https://msft.it/6040vsvEq
Microsoft Research
Think Tanks
Redmond, Washington 406,104 followers
We advance science and technology to benefit humanity.
About us
At Microsoft Research, we accelerate scientific discovery and technology innovation to empower every person and organization on the planet to achieve more. We do this by bringing together the best minds across diverse disciplines and backgrounds to take on the most pressing research challenges for Microsoft and for society. Our Research Lens We consider research directions through the lens of the positive impact we aspire to create with and for customers, communities, and all of society.
- Website
-
http://www.microsoft.com/research
External link for Microsoft Research
- Industry
- Think Tanks
- Company size
- 1,001-5,000 employees
- Headquarters
- Redmond, Washington
- Founded
- 1991
Updates
-
AI agents can't remember past conversations. They must constantly reload or retrieve context, which grows less efficient as tasks get longer and more complex. Memora solves this with a scalable memory system separating what’s stored from how it's retrieved: https://msft.it/6049vs3gh
-
-
What do people actually do with AI at work? A new analysis of five million M365 Copilot conversations has answers. Scott Counts breaks it down in a new video. And dive into the analysis here: https://msft.it/6044vUHsC
-
Microsoft Research reposted this
In the Bay Area, The Garage has reopened with refreshed spaces, new experiences, and even more opportunities for team members to learn, build, and create together. Here’s a look at the moments that made the week one to remember.
-
Researchers introduce generative causal testing, which translates black box models into clear hypotheses and verifies them in the scanner, revealing what specific brain regions respond to in language. https://msft.it/6042vUtRe
-
Talos was built to help resolve a major bottleneck in genomic medicine: human review time. The open-source system recovered 90% of in-scope diagnoses while surfacing just 1.3 candidate variants per patient for expert review. https://msft.it/6045vUBwz
-
On the Microsoft Research Podcast, Microsoft’s Jaime Teevan, Jenna Butler, Jake Hofman, and Rebecca Janssen explore how looking at AI as more than a tool for automating tasks people are already capable of might open the door to new possibilities. https://msft.it/6043vo8cZ
-
Can you remember a new password after listening to five hours of stories? A transformer can. AI researchers Subutai A. and Nicolo Fusi explore the fundamental differences between human memory and machine intelligence with Microsoft CVP Doug Burger. https://msft.it/6047vllRx
-
ResNet has received the Longuet-Higgins Prize at CVPR 2026, recognizing research with proven, lasting impact. A decade after its publication, residual connections remain foundational to how modern AI systems are built, with over 320,000 citations and growing. https://msft.it/6049vl6bW
-
-
Microsoft Research reposted this
I'm delighted to share that our ACM Mobicom 2003 paper titled "Impact of Interference on Multi-hop Wireless Network Performance" that I co-authored with Kamal Jain, Jitendra Padhye, and Lili Qiu has been selected for the ACM SIGMOBILE Test-of-Time Paper Award for 2026 (https://lnkd.in/g-U5-HaP). My co-authors and I are grateful to the award committee and to the SIGMOBILE community for the recognition of this work, which was done during the heyday of wireless mesh networking research at Microsoft Research and beyond. It is particularly memorable for me because of the elegant blending of theory and practice that it represents. This is the 5th test-of-time award I have been fortunate to have received across various ACM SIGs and conferences over my career. At 23 years, this is by far the longest it has been from the time of publication (2003) until the award (2026). This makes the recognition especially satisfying but also a test-of-memory award of sorts 😀