June 23-25, 2026
UC San Diego

A three-day workshop exploring the intersection of artificial intelligence and multimodal scientific research

About AIMS26

AIMS2026 (AI for Multimodal Science) is a three-day interdisciplinary workshop that brings together Schmidt Science Fellows from around the world and UC San Diego postdoctoral researchers to explore how multimodal artificial intelligence can accelerate discovery across scientific domains.

Multimodal AI sits at the forefront of modern research, enabling the integration of heterogeneous data types—such as images, time series, text, graphs, and simulations—to address complex scientific questions. These methods cut across biology, medicine, climate science, physics, materials science, and computer vision, where insights increasingly emerge from the fusion of multiple data modalities rather than from any single source alone.

While scientific datasets are often domain-specific, the core principles of multimodal fusion are broadly transferable. AIMS2026 emphasizes both data-level multimodality and architecture-level multimodality. By focusing on generalizable fusion patterns and model designs, the workshop equips participants with approaches that can be readily adapted across disciplines—simply by swapping in their own domain's data, tasks, and evaluation metrics.

Through a combination of lectures, hands-on sessions, and collaborative discussions, AIMS2026 aims to foster a shared technical language for multimodal AI in science, build cross-domain connections among early-career researchers, and empower participants to apply state-of-the-art multimodal methods to their own research challenges.

Daily Schedule

8:00 AM - 9:00 AM

Breakfast

Light breakfast, badge and gift distribution

9:00 AM - 12:00 PM

Faculty, Industry, and Lightning Talks

Day 1: Three long-format talks (45 min + 5 min Q&A each), with short breaks.

Days 2 & 3: Two long-format talks, followed by four lightning talks by participants (15 min each).

12:00 PM - 1:00 PM

Lunch

Catered lunch and networking

1:00 PM - 2:30 PM

Demo / Coding Tutorial

Example topics:

  • Comparing single-modality vs. multimodal training results
  • Strategies for data fusion (early, intermediate, late)
  • Handling missing modalities and cross-modal alignment
2:30 PM - 3:00 PM

Afternoon Break

Coffee, tea, juice, and light snacks

3:00 PM - 5:00 PM

Afternoon Block

Hackathon (max ~25 participants) or Alternative Activities (max ~25 participants)

Room layout: Alternative activities will be in Room 15A, whereas Hackathon will be in Room 15B.

5:30 PM - 7:00 PM

Social Events

See details below

Hackathon Overview

Maximum ~25 participants

Purpose: Fun, hands-on, educational; light prizes for all participating teams.

Teams: 4–5 teams (4–5 participants each; max ~25). Teams may self-organize, or organizers can assist.

Resources: Starter datasets with train/test splits, baseline code to lower barriers, and volunteer mentors for troubleshooting.

Objectives: Extend or improve an existing model by:

  • Tuning training parameters
  • Trying alternative loss functions
  • Integrating a new modality into training

Alternative Activities During Hackathon

Maximum ~25 participants

Day 1: Paper Discussion Tables

Bring-Your-Own Paper (participant-driven): Each participant selects a multimodal AI–relevant paper, drafts brief notes/questions (e.g., on fusion strategies, evaluation, failure modes), and shares at round tables. A moderator (organizer) ensures balanced discussion.

Day 2: Grant & Fellowship Writing — From Specific Aims to Storytelling

Led by a professional writing coach. Structured exercises, peer feedback, and short coaching rounds help participants draft or revise sections of proposals and research statements, with optional prompts focused on multimodal AI research (e.g., framing interdisciplinary datasets, highlighting integration of methods, or articulating broader impacts).

Day 3: Interactive Q&A Panel

Participants draw prepared question cards with prompts on multimodal systems (e.g., reliability, architectures, applications). Each participant gives a brief response, followed by moderator input with accurate context and examples. This keeps everyone engaged, sparks peer learning, and ensures valuable takeaways.

Social Events

Day 1 (5:30–7:00 PM): UCSD Sandbox Tour

Step inside the Goeddel Family Technology Sandbox and see state-of-the-art imaging and multimodal AI workflows in action. This tour is designed to spark cross‑university and cross‑department collaboration by showcasing shared infrastructure, open research challenges, and concrete pathways for joint multimodal systems projects. Refreshments will be served.

Day 2 (5:30–7:00 PM): Sunset Walk and Ice Cream

Informal sunset walk and ice cream at La Jolla beach.

Day 3 (5:30–7:00 PM): Awards Ceremony and Dinner Reception

The final day of AIMS2026 will conclude with a hackathon awards ceremony, featuring dynamic short demos, results presentations, and live audience voting, followed by the presentation of awards and special gifts. The workshop will then close with an evening dinner reception, creating a welcoming and celebratory atmosphere for participants to network, exchange ideas, and strengthen cross-disciplinary connections formed throughout the workshop.

Registration

Register using the official AIMS 2026 registration form.

Register Now

Speakers

Distinguished researchers and practitioners in AI and multimodal science

Day 1 Speakers

Speaker 1

To be determined

Speaker 2

To be determined

Speaker 3

To be determined

Day 2 Speakers

Speaker 4

To be determined

Speaker 5

To be determined

Day 3 Speakers

Speaker 6

To be determined

Speaker 7

To be determined

Venue

UC San Diego

University of California, San Diego
Seventh College – Tower West, 15th Floor, Room 15A
10176 Scholars Drive
La Jolla, CA 92093
USA

The workshop will be held at UC San Diego, one of the world's leading public research universities. The campus provides state-of-the-art facilities and a vibrant academic environment for interdisciplinary collaboration.

Visit UCSD Website View on Map

Virtual Tour of Room 15A

Explore the workshop venue with our interactive 360° virtual tour

Organizing Committee

Yasmin Kassim

Yasmin Kassim

Executive Lead Organizer
Scientific Program, Workshop Vision & Overall Coordination
Schmidt AI in Science Postdoctoral Fellow
Co-Vice Chair of Exposure to Industry Program, PDA
University of California, San Diego

Biography
Uri Manor

Uri Manor

Senior Scientific Advisor, Faculty Engagement & Partnerships Lead
Assistant Professor
Dr. David V. Goeddel Chancellor’s Endowed Chair in Biological Sciences
University of California, San Diego

Biography
Xiaoyu Zhao

Xiaoyu Zhao

Communications & Registration Lead
Schmidt AI in Science Postdoctoral Fellow
University of California, San Diego

Biography
Eleonora Rachtman

Eleonora Rachtman

Tutorials & Materials Lead
Schmidt AI in Science Postdoctoral Fellow
University of California, San Diego

Biography
Konstantinos Polyzos

Konstantinos Polyzos

Hackathon & Data Lead
Schmidt AI in Science Postdoctoral Fellow
University of California, San Diego

Biography

Tommie Velasquez

Event Logistics & Administrative Coordination
University of California, San Diego Staff

Funding

This workshop is sponsored by Schmidt Sciences in partnership with University of California San Diego.

Institutional Support

This workshop is supported by the UC San Diego Postdoctoral Association (PDA) Executive Board and the Goeddel Family Technology Sandbox.

×

Yasmin Kassim

Yasmin Kassim is a Schmidt AI in Science Fellow and postdoctoral researcher at UC San Diego specializing in biomedical image processing, computer vision, and deep learning. She previously worked at the NIH and Akoya Biosciences and earned a PhD in Computer Engineering from the University of Missouri–Columbia, receiving the 2017 Outstanding PhD Student Award (EECS). Across academia and industry, she builds rigorous AI tools and foundation models that convert images into quantitative insight and scalable measurements, reflected in awards and a strong publication record. Looking ahead with Schmidt, she will channel this momentum into next-generation AI models and tools for organelle analysis and dynamics. Dr. Kassim's STEM Mentor is Uri Manor, and her AI Co-Mentor is Marc Niethammer.

×

Uri Manor

Uri Manor received his Ph.D. with Dr. Bechara Kachar in the Cellular, Molecular, Developmental Biology and Biophysics program at Johns Hopkins University-NIH Graduate Partnership Program and conducted his postdoctoral work in the Lippincott-Schwartz lab at the NIH. He then joined the Salk Institute for Biological Studies as Director of the Waitt Advanced Biophotonics Core in 2016, where he was the recipient of the Chan-Zuckerberg Imaging Scientist Award. He then joined the UC San Diego faculty in 2023 as an Assistant Professor and Director of the Goeddel Family Technology Sandbox.

×

Xiaoyu Zhao

Xiaoyu Zhao completed her PhD in cancer genomics at Stony Brook University. As a postdoctoral scholar in the Department of Medicine at UC San Diego, she is deeply interested in understanding the complex biological systems and human diseases by integrating genome editing technologies and artificial intelligence (AI) models. Her current focus is on developing deep learning models aimed at improving variant interpretation in the context of drug responses and disease progression. Dr. Zhao's STEM Mentor is Trey Ideker and her AI Co-Mentor is Vineet Bafna.

×

Eleonora Rachtman

Eleonora Rachtman earned her PhD in Bioinformatics and Systems Biology from UC San Diego and is now a postdoctoral scholar in Electrical and Computer Engineering. She develops algorithms to analyze big genomic datasets and infer evolutionary relationships between species. Her postdoctoral work focuses on incorporating deep learning into phylogenomic analysis, specifically by developing machine learning algorithms to add new species to large phylogenies without reconstructing them from scratch. This approach enhances existing binning and species identification techniques and aims to address challenges in bacterial and viral metagenomics, improving pathogen detection, understanding microbial diversity, and unraveling evolutionary dynamics of various species. Dr. Rachtman's STEM and AI Mentors are Siavash Mirarab & Davey Smith.

×

Konstantinos D. Polyzos

Konstantinos D. Polyzos is a Postdoctoral Fellow at the Electrical and Computer Engineering department (ECE) at University of California San Diego having recently obtained his Ph.D at the ECE department at the University of Minnesota. His research focuses on learning, inferring and optimizing with just a few data. Specifically, he has been developing and leveraging active-, transfer-, and self-supervised learning and Bayesian optimization methods to learn and/or optimize when only a few input-output data are available due to privacy concerns or high sampling costs, with application to healthcare, 5G networks and robotics. Dr. Polyzos' STEM and AI Mentor is Tara Javidi.