UF Data Studio @ ICDM25

I had the greatest pleasure to attend the IEEE International Conference on Data Mining 2025 (ICDM25) and present my work at the 5th IEEE International Workshop on Multimodal AI (MMAI). This will be the first in a series of posts detailing the research various data scientists and I presented at the conference and its workshops. We at the UF Data Studio hope that these posts inspire readers to check out the groundbreaking work published at the conference and contact us and the other authors with feedback, questions, or potential research opportunities.

The Highlights

Posts about the following highlights of my trip will appear on this website.

My Time in Washington

A travel blog about my time in Washington D.C. including the places I visited, the food I ate, and my journey to and from the conference.

Research @ ICDM25

Over 150 papers were accepted into the proceedings of ICDM25 and presented at the Capital Hilton in the heart of D.C. from November 12th, 2025 to November 16th. The papers, workshops and tutorials that I found most inspiring and impressive will get their own blogpost.

Our Contribution

The UF Data Studio's contribution to ICDM25 was OPTiCAL: An Abstract Positional Reasoning Benchmark for Vision Language Models. A post about it will appear to summarize project and how it was received at MMAI and the conference poster session.

OPTiCAL: An Abstract Positional Reasoning Benchmark for Vision Language Models

I attended ICDM25 to present OPTiCAL: An Abstract Positional Reasoning Benchmark for Vision Language Models at MMAI. Below, I summarize the key findings of the work.

Key Findings

• No model is more than 66% accurate at the tasks overall.
• Models often answer 'hexagon' instead of 'pentagon' when the latter was the answer. There were no hexagons in the data.
• We detect significant (p < 0.01) directional performance bias in each of the models benchmarked.

More Info

The official paper should be available to read on the ICDM25 workshop proceedings, which will be released soon on IEEE Explore. Our code and the Shapes30k dataset are available on GitHub. Feel free to reach out to the authors with questions about the work.

We received invaluable feedback on the present OPTiCAL from the community of multimodal AI researchers at MMAI and other colleagues at ICDM25. If you're working on multimodal AI, and VLM evaluation especially, we'd love to hear from you. Please contact me through the channels available on our webpage.

For more information about our research, return to our homepage: ufdatastudio.com.