
Sam Scarpino
@svscarpino
Scientist. Director @Experiential_AI, Faculty @Northeastern, @sfiscience, @uvmcomplexity; Cofounder @globaldothealth. Now @ scarpino.bsky.social
ID: 2332368062
https://scarpino.github.io 07-02-2014 20:25:44
24,24K Tweet
10,10K Followers
1,1K Following

Latest episode of our podcast is out: "What Kind of Intelligence is an LLM?" We talk with two brilliant scholars who have thought a lot about that question: Tomer Ullman and Murray Shanahan I hope you will give it a listen! complexity.simplecast.com/episodes/natur…




Marta C. Gonzalez receiving her Lagrange Prize from ISI Foundation. Well well well deserved!!



Huge congratulations to C. Brandon Ogbunu!!! His work continues to inspire me and demonstrates how someone can thread the needle of being a world-class scientist and world-class science communicator 🔥🔥🔥

Emails reveal gaping holes in surveillance for human bird flu infections. “I have been distressed and depressed by the lack of epidemiologic data and the lack of surveillance,” says a former head HHS ASPR. My latest San Francisco Chronicle KFF Health News gift link shorturl.at/B12Vz



Thrilled to see our work tracking the routes and means of SARS-CoV-2 variant spread in Chile and South America in PNAS Nexus! academic.oup.com/pnasnexus/adva…




Re-upping: Postdoc position (two-to-three years) at Santa Fe Institute on cogsci-inspired AI approaches to multimodal reasoning. Start date flexible: target is Summer / Fall 2025. PhD in AI, ML, CogSci, or related area (by start date) Apply by Nov. 22. santafe.edu/about/jobs/pos…

Latest out: "Bridging the digital divide: mapping Internet connectivity evolution, inequalities, and resilience in six Brazilian cities". Great collaboration with Nicolò Gozzi Niccolò Comini. Grateful to World Bank and Speedtest by Ookla for the support! epjdatascience.springeropen.com/articles/10.11…



Latest out: Comparative Evaluation of Behavioral-Epidemic Models Using COVID-19 Data. Great collaboration with Nicolò Gozzi and Alessandro Vespignani medrxiv.org/content/10.110…


I'm excited to present my work "LLM Unlearning Evaluation using Knowledge Graphs" at the WiML at NeurIPS Conference next week! This work was done at Institute for Experiential AI at Northeastern with my advisor Sam Scarpino. Come checkout my poster at West Ballroom D from 6:30pm-8pm on 10th Dec!

How can network science help us better evaluate what an LLM "knows" and what it does not "know?" If you're attending NeurIPS Conference, you can learn the answer from one of our amazing Institute for Experiential AI at Northeastern masters students, Rushali Mohbe, at the WiML poster session!
