Microsoft Research Blog - Microsoft Research microsoft.com

AutoAdapt: Automated domain adaptation for large language models

microsoft.com·4d

New Future of Work: AI is driving rapid change, uneven benefits

microsoft.com·2w

ADeLe: Predicting and explaining AI performance across tasks

microsoft.com·3w

AsgardBench: A benchmark for visually grounded interactive planning

microsoft.com·4w

GroundedPlanBench: Spatially grounded long-horizon task planning for robot manipulation

microsoft.com·4w

Systematic debugging for AI agents: Introducing the AgentRx framework

microsoft.com·6w

From raw interaction to reusable knowledge: Rethinking memory for AI agents

microsoft.com·6w

Phi-4-reasoning-vision and the lessons of training a multimodal reasoning model

microsoft.com·7w·Hacker News

CORPGEN advances AI agents for real work

microsoft.com·8w

Media Authenticity Methods in Practice: Capabilities, Limitations, and Directions

microsoft.com·9w

Project Silica’s advances in glass storage technology featured in Nature

microsoft.com·9w·Hacker News

Rethinking imitation learning with Predictive Inverse Dynamics Models

microsoft.com·11w

Paza: Introducing automatic speech recognition benchmarks and models for low resource languages

microsoft.com·11w

Log in to enable infinite scrolling