Pinned Tweet
Replying to @patrickgerard_
This is a new way of thinking about user networks: birds of the same feather don't simply tweet together; they *think* together. Connecting users via latent narratives lets us model with less data and break from arbitrary platform boundaries.
Paper here: arxiv.org/abs/2505.21729
Had a great time last week at the Oxford Internet Institute’s (@oiioxford) Future of Social Media Research workshop!
I presented my work on cross-platform discourse networks — moving past platform silos to a unified view of the online information ecosystem.
Patrick Gerard retweeted
Grateful for the opportunity to speak yesterday at the Oxford Internet Institute during the Future of Social Media Research workshop!
I presented Narrative Networks, my system for using transformer models, multilingual embeddings, and network analysis to track ipropaganda!
Patrick Gerard retweeted
✨ Very overdue update:
I'll be starting as an Assistant Professor in CS at University of Minnesota, Twin Cities, Fall 2026. I will be recruiting PhD students!!
Please help me spread the word! [Thread] 1/n
Patrick Gerard retweeted
Similar trajectory with COVID lab-leak/Fauci narratives—claims about NIH gain-of-function research and intentional virus creation.
Telegram ⚪ + Truth Social 🟣 dominate early and throughout Later gains ground on Twitter 🔵 and TikTok 🔴
Very similar pattern :0
New paper + interactive dashboard on the 2024 US election information ecosystem.
Building on the original discourse networks work with @Hans_Hanley, @LucaLuceri, @emilio__ferrara, we can understand+visualize the online landscape as a unified system across platforms.
When emergence occurs, users tend to have +290% more active cross-platform neighbors discussing the narrative than when it doesn't.
This proximity signal makes cross-platform prediction tractable—no diffusion modeling required.
Take a look for yourself!
🎛️ Dashboard: patrikgerard.github.io/cross…
📄 New Paper: arxiv.org/abs/2510.09464
🌐My Website here: patrickgerard.co
Really curious what patterns you find when exploring the data







