Microsoft Research AI Testing and Evaluation: Learnings from Science and Industry Listen to the series Featured Blogs Aug 20 MindJourney enables AI to explore simulated 3D worlds to improve spatial interpretation Publication Video Featured Blogs Aug 12 Dion: the distributed orthonormal update revolution is here Publication Featured Podcasts Aug 7 Reimagining healthcare delivery and public health with AI Download Publication Featured Blogs Aug 6 Self-adaptive reasoning for science Publication Featured Blogs Aug 5 Project Ire autonomously identifies malware at scale Download Publications Aug 13 Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning Events Sep 8 SIGCOMM 2025 Publication Publications Aug 12 OdysseyBench: Evaluating LLM Agents on Long-Horizon Complex Office Application Workflows Publications Aug 1 RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination Articles Aug 14 StreamMind: AI system that responds to video in real time Publications Aug 1 Collective Agency in Art-making: Towards Community-centric Design of Text-to-Image (T2I) AI Tools Publications Aug 11 Learning from other domains to advance AI evaluation and testing Publications Jul 24 Towards reliable use of artificial intelligence to classify otitis media using otoscopic images: Addressing bias and improving data quality Publications Aug 5 Velocitune: A Velocity-based Dynamic Domain Reweighting Method for Continual Pre-training Blogs Aug 5 VeriTrail: Detecting hallucination and tracing provenance in multi-step AI workflows Publication Video Loading Opens in a new tab
Featured Blogs Aug 20 MindJourney enables AI to explore simulated 3D worlds to improve spatial interpretation Publication Video
Featured Podcasts Aug 7 Reimagining healthcare delivery and public health with AI Download Publication
Publications Aug 13 Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning
Publications Aug 12 OdysseyBench: Evaluating LLM Agents on Long-Horizon Complex Office Application Workflows
Publications Aug 1 RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination
Publications Aug 1 Collective Agency in Art-making: Towards Community-centric Design of Text-to-Image (T2I) AI Tools
Publications Jul 24 Towards reliable use of artificial intelligence to classify otitis media using otoscopic images: Addressing bias and improving data quality
Publications Aug 5 Velocitune: A Velocity-based Dynamic Domain Reweighting Method for Continual Pre-training
Blogs Aug 5 VeriTrail: Detecting hallucination and tracing provenance in multi-step AI workflows Publication Video