Rogerio Bonatti

Rogerio Bonatti is a senior researcher in the Applied Sciences Group (ASG) at Microsoft.  His work focuses on multimodal foundational models for decision-making.  He creates generative machine-learning models that fuse language, vision, and other features to allow systems to take the best actions over time.  Much of his past work before joining ASG was in the robotics space, where he deployed autonomous systems in multiple manipulators, virtual and real embodied agents, all the way to flying robots.

Rogerio received his Ph.D. in robotics from Carnegie Mellon University’s School of Computer Science.  His work has been awarded with a Best Student Paper Finalist Nomination (IROS 2020), a Microsoft Research Dissertation Grant, a Siebel Scholarship and a Swartz Entrepreneurship Fellowship.

He was born and raised in São Paulo, Brazil, and holds a B.S. in mechatronics engineering from the University of São Paulo.

• Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale

  NeurIPS Workshop 2024 December, 2024

  By Rogerio Bonatti, Danny Zhao, Francesco Bonacci, Dillon Dupont, Sara Abdali, Yinheng Li, Justin Wagle, Kazuhito Koishida, Arthur Bucker, Lawrence Jang, Zack Hui
• VideoWebArena: Evaluating Long Context Multimodal Agents with Video Understanding Web Tasks

  NeurIPS Workshop 2024 December, 2024

  By Lawrence Jang, Yinheng Li, Charles Ding, Justin Lin, Paul Pu Liang, Danny Zhao, Rogerio Bonatti, Kazuhito Koishida
• Data Generation using Large Language Models for Text Classification: An Empirical Case Study

  DMLR Workshop in ICML 2024 July, 2024

  By Yinheng Li, Rogerio Bonatti, Sara Abdali, Justin Wagle, Kazuhito Koishida