New version of our preprint! More about the project and data access on our website migrate.tech.cornell.edu

And if you think your work would be inspiring and interesting for this interdisciplinary audience, you might be a good guest speaker. We are still finalizing this semester's schedule, so send me a message!

We bring together PhD students, postdocs, faculty, and practitioners from many disciplines to focus on equity, access, and the role of data in shaping urban spaces. We have biweekly meetings with guest speakers, moderated paper discussions, and hands-on workshops.

Come join us!

Are you a researcher using computational methods to understand cities?

@mfranchi.bsky.social @jennahgosciak.bsky.social and I organize an EAAMO Bridges working group on Urban Data Science and we are looking for new members!

Fill the interest form on our page: urban-data-science-eaamo.github.io

*Proud advisor moment* My (first) PhD student Zhi Liu (zhiliu724.github.io) is 1 of 4 finalists for the INFORMS Dantzig Dissertation Award, the premier dissertation award for the OR community. His dissertation spanned work with 2 NYC govt agencies, on measuring and mitigating operational inequities

"Removing the protected bike lane won’t remove cyclists — it will only make the street less safe," the Department of Transportation said in new testimony.

"The city risks legal liability for knowingly reducing safety on a Vision Zero priority corridor."
buff.ly/QNgRyts

New work 🎉: conformal classifiers return sets of classes for each example, with a probabilistic guarantee the true class is included. But these sets can be too large to be useful.

In our #CVPR2025 paper, we propose a method to make them more compact without sacrificing coverage.

I’m really excited to share the first paper of my PhD, “Learning Disease Progression Models That Capture Health Disparities” (accepted at #CHIL2025)! ✨ 1/

📄: arxiv.org/abs/2412.16406

Can vision-language models understand figurative meaning in multimodal inputs, like visual metaphors, sarcastic captions or memes? Come find out at our #NAACL2025 poster on Friday at 9am!

New task & dataset of images and captions with figurative phenomena like metaphor, idiom, sarcasm, and humor.

I became a dog scientist on April 1st. Now back to normal (a cat scientist).

You can request data access on our website! This is joint work with Rachel Young, Maria Fitzpatrick, @nkgarg.bsky.social, and @emmapierson.bsky.social.

Paper: arxiv.org/abs/2503.20989
Website: migrate.tech.cornell.edu
5/5

MIGRATE reveals trends (like dramatic rates of out-migration in response to California wildfires) that are entirely invisible in public data.

This case study only scratches the surface of what’s possible with this data–we’re excited to see what you do with it!

4/5

We produce MIGRATE by developing an iterative-proportional-fitting based algorithm to reconcile (1) granular but biased proprietary data and (2) coarser but more reliable Census data.

We comprehensively validate MIGRATE against external data sources.

3/5

MIGRATE is:

Thousands of times more granular than existing public migration datasets
Highly correlated with external Census datasets
Less biased, and more consistent with Census data, than proprietary address data

2/5

Migration data lets us study responses to environmental disasters, social change patterns, policy impacts, etc. But public data is too coarse, obscuring these important phenomena!

We build MIGRATE: a dataset of yearly flows between 47 billion pairs of US Census Block Groups. 1/5

💡New preprint & Python package: We use sparse autoencoders to generate hypotheses from large text datasets.

Our method, HypotheSAEs, produces interpretable text features that predict a target variable, e.g. features in news headlines that predict engagement. 🧵1/