Sophisticated side-channel attacks present new security challenges for cloud providers. Learn how Microsoft is exploring defenses against emerging attacks with principled microarchitectural isolation: https://lnkd.in/e-qDU5qJ
Microsoft Research
Think Tanks
Redmond, Washington 296,653 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
-
In “Collaborators,” Siddharth Suri and David Holtz give a brief history of prompt engineering, discuss the debate behind their recent collaboration, and share what they found from studying how people’s approaches to prompting change as models advance. https://lnkd.in/eVXRBACN
-
-
Microsoft Research’s AI2BMD, an AI-based system that efficiently simulates a wide range of proteins in all-atom resolution, can advance drug discovery and biomolecular research. https://lnkd.in/es2c2eUr
-
-
The Rust verification tool Verus aims to make formal verification more widely available to developers, helping them create software that behaves as intended. Learn about the work, an award winner at SOSP ’24, w/ Chris Hawblitzel & Jay Lorch on “Abstracts”: https://lnkd.in/e8xdxtk9
-
-
A new toolkit using LLMs, combined w/ fault injection & repurposed unit tests, can detect common “retry” bugs in software systems, helping improve their robustness. Learn more about the work, featured at SOSP ’24, on “Abstracts” w/ Shan Lu & Bogdan Stoica: https://lnkd.in/ex29SBqz
-
-
Award-winning research for building resilient systems, as well as scaling deep learning computation & reproducing failures in production are just some ways Microsoft researchers are advancing the state of the art in computer systems research at SOSP 24: https://lnkd.in/efDGkXDE
-
-
When augmented with AI, small power grids can create opportunities for decentralized, equitable, and resilient power. Microsoft’s collaboration shows AI’s potential to optimize energy distribution and empower communities: https://lnkd.in/ekSDUH62
-
-
New Research | FLASH: Workflow automation agent for diagnosing recurring incidents; METAREFLECTION: Learning instructions for language agents using past reflections; Boosting LLM training efficiency through faster communication between GPUs; and more: https://lnkd.in/enQ88Bc8
-
-
GraphRAG leverages semantic structuring of data to generate responses to complex user queries. A collaboration with Uncharted expands the frontiers of this technology, developing a new approach to processing local queries: DRIFT Search https://lnkd.in/ePhGs54K
-
-
Microsoft Research reposted this
Strategy, Storytelling, Research and Thought Leadership for Azure AI at Microsoft | Keynote Speaker
I had the opportunity to interview Ade Famoti, Head of Research Incubations at Microsoft Research Accelerator, at the Gartner IT Symposium/Xpo last week. We talked about the technology paradigm shift we’re living through, and how he believes it differs in scale and impact from what we saw with the PC, Internet, mobile and cloud computing eras. Here are a few tidbits from our conversation: On multimodality: “Multimodality is one of the inflection points of this generative AI moment, in this AI renaissance we're in.” One example, he says, is LLaVA Med, a multimodal-language model from Microsoft that is designed to extract valuable insights from complex biomedical images and engage in meaningful conversation with users. Another example is GigaPath, a multimodal whole-slide foundational model aimed at tackling the complexities of digital pathology. “The point,” he says, “is not to replace the human, but to augment the human.” On embodied AI: “Some folks would call that robotics; it’s not just robotics, but the spectrum of virtual agents and physical agents that sense, perceive and act in real world environments” Why? Because robotics has been largely rules-based, but “when you throw those kinds of embodiments into environments that have uncertainty, they start to flail.” Embodied AI, however, represents the convergence of AI with Newtonian physics, which is enabling us to solve 70-year-old robotics problems such as the ability to navigate with uncertainty and to bring optimization to how robots manipulate, locomote, and perceive the world. On AI for scientific discovery: LLMs or generative AI models can help us mirror the physical laws of nature with exquisite precision, such as thermodynamics or Schrödinger’s wave function equations (Schrödinger’s equation describes the behavior of molecules and materials at the subatomic level), so we can build deep learning models based upon them. We can then use these models to generate novel materials for scientific discovery; for example, lithium ion battery electrolytes for mobile devices or electric vehicles, or deep learning models for drug discovery pipelines and to understand protein synthesis. On the platform implications of this AI paradigm shift: “To get the advantages and impact of everything we’ve discussed—multimodal models, embodied AI, scientific discovery—into the hands of end users, you have to imagine infrastructures that can support that. Are you investing in platforms that have the versatility, agility, security, computational and data infrastructures that meet your needs? Invest in an infrastructure player that’s ready now to support your computational needs--so you can deliver new outcomes and new value to your customers in the future.” #Azure #MicrosoftResearch #MultimodalAI #EmbodiedAI #LLMs #GenerativeAI #GartnerSym
-