Symposium spotlights AI's potential to revolutionize health care Robot-assisted surgery, an app for diagnosing strep throat, and a new tool to detect glaucoma are all on the health care horizon thanks to artificial intelligence innovators at Johns Hopkins Universit...

From work and education to the services we rely on in homes, AI has rapidly become a ubiquitous part of our everyday lives," said #HopkinsDSAI Interim Director Rama Chellappa, at last week's Johns Hopkins Data Science and AI Institute Fall 2025 Symposium.

hub.jhu.edu/2025/09/24/a...

The Johns Hopkins Data Science and AI Institute Fall 2025 Symposium kicked off with remarks from Rama Chellappa, interim director of #HopkinsDSAI. Benjamin Lee of the University of Pennsylvania is giving his talk "Toward Sustainable Data Centers for Artificial Intelligence."

Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life - Johns Hopkins Data Science and AI Institute Join us for the Johns Hopkins Data Science and AI Institute Fall 2025 Symposium AI in Daily Life Tuesday, Sept. 16 Symposium | 1 to 4:15 p.m. | Shriver Hall Auditorium…

The Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life is on Tuesday, September 16. The event will explore how #AI is reshaping everyday life.

Event details and registration:
ai.jhu.edu/event/fall-2...

Our pioneering partnership with leading cancer centers and tech companies has been recognized by TIME magazine’s #TIME100AI 2025 for groundbreaking work in artificial intelligence and cancer research!

Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life - Johns Hopkins Data Science and AI Institute Join us for the Johns Hopkins Data Science and AI Institute Fall 2025 Symposium AI in Daily Life Tuesday, Sept. 16 Symposium | 1 to 4:15 p.m. | Shriver Hall Auditorium…

Register to attend the Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life on Tuesday, September 16. The event will explore how #AI is reshaping everyday life.

Event details: ai.jhu.edu/event/fall-2...

#HopkinsDSAI is proud to announce its inaugural postdoctoral fellows for the 2025-2026 academic year. These distinguished scholars join our growing research community, collaborating with faculty across a wide range of academic disciplines and divisions.

ai.jhu.edu/news/johns-h...

#HopkinsDSAI welcomes 22 new faculty members, who join more than 150 DSAI faculty members across @jhu.edu in advancing the study of data science, machine learning, and #AI and translation to a range of critical and emerging fields.

ai.jhu.edu/news/data-sc...

Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life - Johns Hopkins Data Science and AI Institute Join us for the Johns Hopkins Data Science and AI Institute Fall 2025 Symposium AI in Daily Life Tuesday, Sept. 16 Symposium | 1 to 4:15 p.m. | Shriver Hall Auditorium…

The Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life on Tuesday, September 16. The event will explore how #AI is reshaping everyday life. Register to attend: ai.jhu.edu/event/fall-2...

“Currently, for every 20 defibrillators implanted, only one patient derives benefit." @ntrayanova.bsky.social talks about using AI to transform how we prevent sudden cardiac death on
KPCW's Cool Science Radio: www.kpcw.org/show/cool-sc...

In the world of AI, who speaks for the trees? Johns Hopkins computer scientists’ new training method, EigenLoRAx, promises to cut the environmental cost of fine-tuning large AI models.

Johns Hopkins computer scientists have developed a method that dramatically cuts the environmental costs of fine-tuning #AI. They call it EigenLoRAx and, like the Dr. Seuss character, it’s a nod to a greener future. 🌳

Talking robots learn to manage human interruptions Johns Hopkins computer scientists designed an interruption-handling system to facilitate more natural conversations with social robots

Johns Hopkins researchers including @victor-antony.bsky.social and @ziangxiao.bsky.social‬ have created a system that could make social robots more effective at detecting and managing user interruptions—a breakthrough for areas like health care and education where natural conversation is crucial.

Register to attend the Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life on Tuesday, September 16. The event will explore how #AI is reshaping everyday life.

Learn more here: ai.jhu.edu/event/fall-2...

AI advancements: What congressional staffers need to know Johns Hopkins AI experts share the main things congressional staffers should know about AI's impact on national security, ethics, and more.

Last month, @jhu.edu faculty members hosted an information session on the latest developments in #AI and machine learning at the Johns Hopkins University Bloomberg Center for congressional staffers.

Read more about the event: washingtondc.jhu.edu/news/ai-cong...

From 'woke' AI to data centers, can Trump's plan help America win the AI race? The White House released its action plan to help America win the race for artificial intelligence innovation.

IAA's Anton Dahbura spoke with The National Desk about the need for safety guardrails to address emerging #AI applications.

thenationaldesk.com/news/america...

Exciting informational session on AI for congressional staff today at @jhu.edu Bloomberg Center in coordination with #HopkinsDSAI, Johns Hopkins Engineering Lifelong Learning, and Office of Federal Strategy.

Register today to attend the Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life on Tuesday, September 16. The event will explore how #AI is reshaping everyday life.

Details available here: ai.jhu.edu/event/fall-2...

Multimodal AI to forecast arrhythmic death in hypertrophic cardiomyopathy Nature Cardiovascular Research - Lai et al. present the machine learning model MAARS to predict arrhythmic sudden cardiac death from multimodal imaging and clinical data in patients with...

So thrilled! Our paper is out in Nature Cardiovascular Research! We developed #multimodal #AI to predict #cardiacarrest in hyperthrophic #cardiomyopathy. The algorithm significantly outperformed clinical guidelines!
@hopkinsengineer.bsky.social

rdcu.be/euv0q

AI Predicts Patients Likely to Die of Sudden Cardiac Arrest - Johns Hopkins Whiting School of Engineering Model much better than doctors at identifying patients most at-risk

In @natcardiovascres.nature.com: One of cardiology’s hardest problems is IDing which patients with hypertrophic cardiomyopathy are at risk of sudden cardiac death.
@ntrayanova.bsky.social's new AI model achieved 89% accuracy, offering a path for prevention. engineering.jhu.edu/news/ai-pred...

Robot Performs First Realistic Surgery Without Human Help - Johns Hopkins Whiting School of Engineering System trained on videos of surgeries performs like an expert surgeon

Realistic robotic surgery without human help! Assoc Prof Axel Krieger's SRT-H performed a lengthy phase of a gallbladder removal w/ the expertise of a skilled human surgeon—even during unexpected scenarios that could cause emergencies. engineering.jhu.edu/news/robot-p...

Save the date for the Johns Hopkins Data Science and AI Institute Fall 2025 Symposium: AI in Daily Life on Tuesday, September 16!

Additional information and registration details coming soon. #HopkinsDSAI

ai.jhu.edu/event/johns-...

Meet #HopkinsDSAI at booth 1317 of the IEEE / CVF Computer Vision and Pattern Recognition Conference! See today's booth schedule below. #CVPR2025

Friday, June 13
11 a.m. Tinoosh Mohsenin
12 p.m. Alan Yuille
4:30 p.m. Reno Kriz
5:30 p.m. Rama Chellappa

Hopkins researchers including JHU Electrical and Computer Engineering's Tinoosh Mohsenin and @jhu-bdps.bsky.social Rama Chellappa are speaking at booth 1317 of the IEEE / CVF Computer Vision and Pattern Recognition Conference today!

Come meet #HopkinsDSAI #CVPR2025

Meet #HopkinsDSAI at booth 1317 of the IEEE / CVF Computer Vision and Pattern Recognition Conference this week! #CVPR2025

SpatialLLM: A Compound 3D-Informed Design towards Spatially-Intelligent Large Multimodal Models Humans naturally understand 3D spatial relationships, enabling complex reasoning like predicting collisions of vehicles from different directions. Current large multimodal models (LMMs), however, lack...

A Compound 3D-Informed Design Toward Spatially-Intelligent Large Multimodal Models
Wufei Ma, Luoxin Ye, Nessa McWeeney, Celso M de Melo, Alan Yuille, and Jieneng Chen
www.arxiv.org/abs/2505.00788

Adventurer: Optimizing Vision Mamba Architecture Designs for Efficiency In this work, we introduce the Adventurer series models where we treat images as sequences of patch tokens and employ uni-directional language models to learn visual representations. This modeling par...

Adventurer: Optimizing Vision Mamba Architecture Designs for Efficiency
Feng Wang, Timing Yang, Yaodong Yu, Sucheng Ren, Guoyizhe Wei, Angtian Wang, Wei Shao, Yuyin Zhou, Alan Yuille, and Cihang Xie
arxiv.org/abs/2410.07599

CamFreeDiff: Camera-free Image to Panorama Generation with Diffusion Model This paper introduces Camera-free Diffusion (CamFreeDiff) model for 360-degree image outpainting from a single camera-free image and text description. This method distinguishes itself from existing st...

CamFreeDiff: Camera-Free Image to Panorama Generation With Diffusion Model
Xiaoding Yuan, Shitao Tang, Kejie Li, Alan Yuille, and Peng Wang
arxiv.org/abs/2407.07174

Distilling Multi-modal Large Language Models for Autonomous Driving Autonomous driving demands safe motion planning, especially in critical "long-tail" scenarios. Recent end-to-end autonomous driving systems leverage large language models (LLMs) as planners to improve...

Distilling Multi-Modal Large Language Models for Autonomous Driving
Deepti Hegde, Rajeev Yasarla, Hong Cai, Shizhong Han, Apratim Bhattacharyya, Shweta Mahajan, Litian Liu, Risheek Garrepalli, Vishal M. Patel, and Fatih Porikli
arxiv.org/abs/2501.09757

Filter Images First, Generate Instructions Later: Pre-Instruction Data Selection for Visual Instruction Tuning Visual instruction tuning (VIT) for large vision-language models (LVLMs) requires training on expansive datasets of image-instruction pairs, which can be costly. Recent efforts in VIT data selection a...

Filter Images First, Generate Instructions Later: Pre-Instruction Data Selection for Visual Instruction Tuning
Bardia Safaei, Faizan Siddiqui, Jiacong Xu, Vishal M. Patel, and Shao-Yuan Lo
arxiv.org/abs/2503.07591

Flowing from Words to Pixels: A Noise-Free Framework for Cross-Modality Evolution Diffusion models, and their generalization, flow matching, have had a remarkable impact on the field of media generation. Here, the conventional approach is to learn the complex mapping from a simple ...

Flowing From Words to Pixels: A Noise-Free Framework for Cross-Modality Evolution
Qihao Liu, Xi Yin, Alan Yuille, Andrew Brown, and Mannat Singh
arxiv.org/abs/2412.15213

GenDeg: Diffusion-based Degradation Synthesis for Generalizable All-In-One Image Restoration Deep learning-based models for All-In-One Image Restoration (AIOR) have achieved significant advancements in recent years. However, their practical applicability is limited by poor generalization to s...

GenDeg: Diffusion-Based Degradation Synthesis for Generalizable All-in-One Image Restoration
Sudarshan Rajagopalan, Nithin Gopalakrishnan Nair, Jay N. Paranjape, and Vishal M. Patel
arxiv.org/abs/2411.17687