Microsoft Research’s Post

Foundation models are fueling a fundamental shift in computing research, natural sciences, social sciences and computing education itself. And as industry-led advances in AI continue to reach new heights, we believe that a vibrant and diverse research ecosystem is essential to realizing the promise of AI to benefit each individual, organization and society worldwide.

Join us on June 27 for the monthly AI, Machine Learning and Computer Vision Meetup. Register for the Zoom: https://lnkd.in/gypQvViN We have three great talks scheduled including: ** Improved Visual Grounding through Self-Consistent Explanations ** Vision-and-language models that are trained to associate images with text have shown to be effective for many tasks, including object detection and image segmentation. In this talk, we will discuss how to enhance vision-and-language models’ ability to localize objects in images by fine-tuning them for self-consistent visual explanations. We propose a method that augments text-image datasets with paraphrases using a large language model and employs SelfEQ, a weakly-supervised strategy that promotes self-consistency in visual explanation maps. This approach broadens the model’s working vocabulary and improves object localization accuracy, as demonstrated by performance gains on competitive benchmarks. About the Speakers Dr. Paola Cascante-Bonillareceived her Ph.D. in Computer Science at Rice University in 2024, advised by Professor Vicente Ordóñez Román, working on Computer Vision, Natural Language Processing, and Machine Learning. She received a Master of Computer Science at the University of Virginia and a B.S. in Engineering at the Tecnológico de Costa Rica. Paola will join Stony Brook University (SUNY) as an Assistant Professor in the Department of Computer Science. Ruozhen (Catherine) HE is a first-year Computer Science PhD student at Rice University, advised by Prof. Vicente Ordóñez, focusing on efficient algorithms in computer vision with less or multimodal supervision. She aims to leverage insights from neuroscience and cognitive psychology to develop interpretable algorithms that achieve human-level intelligence across versatile tasks. #computervision #ai #artificialintelligence #machinevision #machinelearning #datascience #huggingface

Join us on June 27 for the monthly AI, Machine Learning and Computer Vision Meetup. Register for the Zoom: https://lnkd.in/gypQvViN We have three great talks scheduled including: ** Improved Visual Grounding through Self-Consistent Explanations ** Vision-and-language models that are trained to associate images with text have shown to be effective for many tasks, including object detection and image segmentation. In this talk, we will discuss how to enhance vision-and-language models’ ability to localize objects in images by fine-tuning them for self-consistent visual explanations. We propose a method that augments text-image datasets with paraphrases using a large language model and employs SelfEQ, a weakly-supervised strategy that promotes self-consistency in visual explanation maps. This approach broadens the model’s working vocabulary and improves object localization accuracy, as demonstrated by performance gains on competitive benchmarks. About the Speakers Dr. Paola Cascante-Bonilla received her Ph.D. in Computer Science at Rice University in 2024, advised by Professor Vicente Ordóñez Román, working on Computer Vision, Natural Language Processing, and Machine Learning. She received a Master of Computer Science at the University of Virginia and a B.S. in Engineering at the Tecnológico de Costa Rica. Paola will join Stony Brook University (SUNY) as an Assistant Professor in the Department of Computer Science. Ruozhen (Catherine) HE is a first-year Computer Science PhD student at Rice University, advised by Prof. Vicente Ordónez-Román, focusing on efficient algorithms in computer vision with less or multimodal supervision. She aims to leverage insights from neuroscience and cognitive psychology to develop interpretable algorithms that achieve human-level intelligence across versatile tasks. #computervision #ai #artificialintelligence #machinevision #machinelearning #datascience #huggingface

Join us on June 27 for the monthly AI, Machine Learning and Computer Vision Meetup. Register for the Zoom: https://lnkd.in/gypQvViN We have three great talks scheduled including: ** Improved Visual Grounding through Self-Consistent Explanations ** Vision-and-language models that are trained to associate images with text have shown to be effective for many tasks, including object detection and image segmentation. In this talk, we will discuss how to enhance vision-and-language models’ ability to localize objects in images by fine-tuning them for self-consistent visual explanations. We propose a method that augments text-image datasets with paraphrases using a large language model and employs SelfEQ, a weakly-supervised strategy that promotes self-consistency in visual explanation maps. This approach broadens the model’s working vocabulary and improves object localization accuracy, as demonstrated by performance gains on competitive benchmarks. About the Speakers Dr. Paola Cascante-Bonilla received her Ph.D. in Computer Science at Rice University in 2024, advised by Professor Vicente Ordóñez Román, working on Computer Vision, Natural Language Processing, and Machine Learning. She received a Master of Computer Science at the University of Virginia and a B.S. in Engineering at the Tecnológico de Costa Rica. Paola will join Stony Brook University (SUNY) as an Assistant Professor in the Department of Computer Science. Ruozhen (Catherine) HE is a first-year Computer Science PhD student at Rice University, advised by Prof. Vicente Ordónez-Román, focusing on efficient algorithms in computer vision with less or multimodal supervision. She aims to leverage insights from neuroscience and cognitive psychology to develop interpretable algorithms that achieve human-level intelligence across versatile tasks. #computervision #ai #artificialintelligence #machinevision #machinelearning #datascience #huggingface

Join us on June 27 for the monthly AI, Machine Learning and Computer Vision Meetup. Register for the Zoom: https://lnkd.in/gypQvViN We have three great talks scheduled including: ** Improved Visual Grounding through Self-Consistent Explanations ** Vision-and-language models that are trained to associate images with text have shown to be effective for many tasks, including object detection and image segmentation. In this talk, we will discuss how to enhance vision-and-language models’ ability to localize objects in images by fine-tuning them for self-consistent visual explanations. We propose a method that augments text-image datasets with paraphrases using a large language model and employs SelfEQ, a weakly-supervised strategy that promotes self-consistency in visual explanation maps. This approach broadens the model’s working vocabulary and improves object localization accuracy, as demonstrated by performance gains on competitive benchmarks. About the Speakers Dr. Paola Cascante-Bonilla received her Ph.D. in Computer Science at Rice University in 2024, advised by Professor Vicente Ordóñez Román, working on Computer Vision, Natural Language Processing, and Machine Learning. She received a Master of Computer Science at the University of Virginia and a B.S. in Engineering at the Tecnológico de Costa Rica. Paola will join Stony Brook University (SUNY) as an Assistant Professor in the Department of Computer Science. Ruozhen (Catherine) HE is a first-year Computer Science PhD student at Rice University, advised by Prof. Vicente Ordóñez, focusing on efficient algorithms in computer vision with less or multimodal supervision. She aims to leverage insights from neuroscience and cognitive psychology to develop interpretable algorithms that achieve human-level intelligence across versatile tasks. #computervision #ai #artificialintelligence #machinevision #machinelearning #datascience #huggingface

Join us on June 27 for the monthly AI, Machine Learning and Computer Vision Meetup. Register for the Zoom: https://lnkd.in/gypQvViN We have three great talks scheduled including: ** Improved Visual Grounding through Self-Consistent Explanations ** Vision-and-language models that are trained to associate images with text have shown to be effective for many tasks, including object detection and image segmentation. In this talk, we will discuss how to enhance vision-and-language models’ ability to localize objects in images by fine-tuning them for self-consistent visual explanations. We propose a method that augments text-image datasets with paraphrases using a large language model and employs SelfEQ, a weakly-supervised strategy that promotes self-consistency in visual explanation maps. This approach broadens the model’s working vocabulary and improves object localization accuracy, as demonstrated by performance gains on competitive benchmarks. About the Speakers Dr. Paola Cascante-Bonillareceived her Ph.D. in Computer Science at Rice University in 2024, advised by Professor Vicente Ordóñez Román, working on Computer Vision, Natural Language Processing, and Machine Learning. She received a Master of Computer Science at the University of Virginia and a B.S. in Engineering at the Tecnológico de Costa Rica. Paola will join Stony Brook University (SUNY) as an Assistant Professor in the Department of Computer Science. Ruozhen (Catherine) HE is a first-year Computer Science PhD student at Rice University, advised by Prof. Vicente Ordóñez, focusing on efficient algorithms in computer vision with less or multimodal supervision. She aims to leverage insights from neuroscience and cognitive psychology to develop interpretable algorithms that achieve human-level intelligence across versatile tasks. #computervision #ai #artificialintelligence #machinevision #machinelearning #datascience #huggingface

Join us on June 27 for the monthly AI, Machine Learning and Computer Vision Meetup. Register for the Zoom: https://lnkd.in/gypQvViN We have three great talks scheduled including: ** Improved Visual Grounding through Self-Consistent Explanations ** Vision-and-language models that are trained to associate images with text have shown to be effective for many tasks, including object detection and image segmentation. In this talk, we will discuss how to enhance vision-and-language models’ ability to localize objects in images by fine-tuning them for self-consistent visual explanations. We propose a method that augments text-image datasets with paraphrases using a large language model and employs SelfEQ, a weakly-supervised strategy that promotes self-consistency in visual explanation maps. This approach broadens the model’s working vocabulary and improves object localization accuracy, as demonstrated by performance gains on competitive benchmarks. About the Speakers Dr. Paola Cascante-Bonilla received her Ph.D. in Computer Science at Rice University in 2024, advised by Professor Vicente Ordóñez Román, working on Computer Vision, Natural Language Processing, and Machine Learning. She received a Master of Computer Science at the University of Virginia and a B.S. in Engineering at the Tecnológico de Costa Rica. Paola will join Stony Brook University (SUNY) as an Assistant Professor in the Department of Computer Science. Ruozhen HE is a first-year Computer Science PhD student at Rice University, advised by Prof. Vicente Ordóñez, focusing on efficient algorithms in computer vision with less or multimodal supervision. She aims to leverage insights from neuroscience and cognitive psychology to develop interpretable algorithms that achieve human-level intelligence across versatile tasks. #computervision #ai #artificialintelligence #machinevision #machinelearning #datascience #huggingface

Join us on June 27 for the monthly AI, Machine Learning and Computer Vision Meetup. Register for the Zoom: https://lnkd.in/gASBEFGa We have three great talks scheduled including: ** Improved Visual Grounding through Self-Consistent Explanations ** Vision-and-language models that are trained to associate images with text have shown to be effective for many tasks, including object detection and image segmentation. In this talk, we will discuss how to enhance vision-and-language models’ ability to localize objects in images by fine-tuning them for self-consistent visual explanations. We propose a method that augments text-image datasets with paraphrases using a large language model and employs SelfEQ, a weakly-supervised strategy that promotes self-consistency in visual explanation maps. This approach broadens the model’s working vocabulary and improves object localization accuracy, as demonstrated by performance gains on competitive benchmarks. About the Speakers Dr. Paola Cascante-Bonilla received her Ph.D. in Computer Science at Rice University in 2024, advised by Professor Vicente Ordóñez Román, working on Computer Vision, Natural Language Processing, and Machine Learning. She received a Master of Computer Science at the University of Virginia and a B.S. in Engineering at the Tecnológico de Costa Rica. Paola will join Stony Brook University (SUNY) as an Assistant Professor in the Department of Computer Science. Ruozhen HE is a first-year Computer Science PhD student at Rice University, advised by Prof. Vicente Ordóñez, focusing on efficient algorithms in computer vision with less or multimodal supervision. She aims to leverage insights from neuroscience and cognitive psychology to develop interpretable algorithms that achieve human-level intelligence across versatile tasks. #computervision #ai #artificialintelligence #machinevision #machinelearning #datascience #huggingface

From Months to Minutes: OpenAI o1 Revolutionizes Scientific Research OpenAI's latest preview model, o1, is making waves in the scientific community with its ability to dramatically accelerate research. This new iteration represents a significant leap forward, especially in fields requiring complex reasoning like quantum physics and molecular biology. Researchers have witnessed o1's impressive capabilities firsthand. Mario Krenn from the Max Planck Institute notes, "In my field of quantum physics, it gives significantly more detailed and coherent responses" compared to previous models. Kyle Kabasares from the Bay Area Environmental Research Institute managed to replicate months of PhD work in just an hour using o1. o1's performance is remarkable: • First AI to outperform PhD scholars on the Graduate-Level Google-Proof Q&A Benchmark (GPQA) • 93% score in physics on the GPQA Diamond test • 83% correct solutions on International Mathematics Olympiad qualifying exam (up from 13% with GPT-4o) o1 uses chain-of-thought reasoning, providing reconstructed summaries of its logic. This approach, coupled with longer learning stages and "thinking" time, contributes to its enhanced capabilities. However, o1 is not without limitations. Anecdotal feedback suggests it may hallucinate more frequently than predecessors. Safety concerns in physical experiments have also been noted. As Andrew White from FutureHouse emphasizes, expert verification remains crucial, especially for novice users. Despite these caveats, o1 is proving invaluable across various scientific domains: • Scanning research and generating new ideas with tools like SciMuse • Accelerating coding tasks with o1-mini • Connecting genetic and patient data in rare disease research As we explore o1's potential, it's clear that AI is becoming an indispensable partner in pushing the boundaries of scientific discovery. However, responsible use and expert oversight remain paramount. What potential applications do you see for o1 in your field? How might we balance innovation with responsible AI use in research? #AI #OpenAI #ScientificResearch #QuantumComputing #Innovation #ResponsibleAI #FutureOfScience https://lnkd.in/g3ysMVsW

Exciting news! 👏 A team of Northeastern University researchers, led by core faculty at our institute David Bau, received a $9 million grant from the National Science Foundation (NSF) for groundbreaking #research to better understand #AIsystems and their societal impacts. “We fundamentally don’t understand how these systems work, what they learned from the data, what their internal algorithms are,” Bau, the lead PI on the National Deep Inference Fabric project, told Northeastern Global News. “I consider it one of the greatest mysteries facing scientists today — what is the basis for synthetic cognition?” Read the full story to learn more about the effort to unlock the secrets of #AI: https://bit.ly/4dAfSDp Khoury College of Computer Sciences #generativeAI #artificialintelligence #LLMs #largelanguagemodels #genAI #AInews #AIresearch #tech #data