Talk & Coverage

Invited Talk

1. M3P: Learning Universal Representations via Multitask Multilingual Multimodal Pre-training, Qingyuan Talk

Media Coverage

1. Microsoft Researchers Propose Open-Vocabulary Responsible Visual Synthesis (ORES) with the Two-Stage Intervention Framework, MarktechPost
2. Microsoft Researchers Propose NUWA-XL: A Novel Diffusion Over Diffusion Architecture For Extremely Long Video Generation, MarktechPost
3. An Undergraduate Student Who Studies NLP Submitted A Paper to A Top CV Conference and Got Accepted as the First Author? (Chinese), AI Tech Review
4. Overcoming Language Barriers! Harbin Institute of Technology, in Collaboration with MSRA, Has Proposed a Unified Pre-training Model for Multi-task, Multi-modal, and Multi-language Applications, M3P (CVPR 2021) (Chinese), FightingCV
5. 16 Descriptions Generate an 11-minute Animation! The New Member of the NUWA Series: the Ultra-long Video Generation Model NUWA-XL (Chinese), AI Era
6. CVPR 2021 | 9 Selected Papers: An Overview of the Latest Advances in Visual Research at Microsoft Research Asia (Chinese), Microsoft Research Asia
7. Microsoft Research Asia’s Multimodal Model NÜWA: Creating Visual Content with Natural Language (Chinese), Microsoft News Centre