Stephen Pfohl

Really great to see this published! So, what is this all about? 🧵 (1/

🎉 We are excited to present two recent works detailing the landscape of AI in health in Africa from a fairness and equity angle: The Nteasee Study arxiv.org/abs/2409.12197 and The Case for Globalizing Fairness dl.acm.org/doi/10.1145/36… (1/11)

Also want to thank the editorial team (@NatureMedicine, @LorenzoRighett7) and reviewers (@MMccradden and anonymous reviewers) for their constructive feedback and support for the work.

This work represents the contributions of many amazing collaborators from @GoogleAI @GoogleDeepMind @MIT @UAlberta, including @hcolelewis @thekaransinghal @DrNealResearch @dr_nyamewaa @adoubleva @weballergy @AziziShekoofeh @negar_rz @LiamGMcCoy @HardyShakerman and many more!

Excited to announce that our paper, “A toolbox surfacing health equity harms and biases in large language models” is now published with @NatureMedicine: nature.com/articles/s4159….

@2plus2make5 @Berkeley_EECS Congrats!

Life update: I'll be joining the @Berkeley_EECS faculty in Jan 2025! I'll also be part of the Computational Precision Health program. I'm excited to continue our work using ML to improve health + social equity at Berkeley, with its history of social justice + public service! 1/

@kdpsinghlab If this works like the pd.eval or pd.query, you could actually drop the f string and replace {value} with @value

👀

@_isupriya I'll share it once the FAccT proceedings are online (not sure when that will be exactly, sorry!)

@stephenpfohl may I get the link of your paper?

Made it to Rio for #FAccT2024

Red teaming LLM’s w @TristanNaumann @MSFTResearch @stephenpfohl @GoogleResearch #SAIL2024 #AI detecting threats across the broad taxonomy of LLM vulnerabilities

As our attendees settle in the @hyattregencypr, we get ready for a deep dive into medical AI with @TristanNaumann and @stephenpfohl presenting “Red Teaming to Test Limitations of LLMs”. Welcome everyone to #SAIL24! We hope you enjoy the conference and beautiful Puerto Rico. 🏝️

Landed in Puerto Rico for #SAIL2024

@davidthewid @FAccTConference @mikarv @struthious @Borhane_B_H @andrewthesmart Great to hear! I was wondering if I needed to set an alarm for 4:30AM to resubmit after the TAPS support gets back to me 😅

@stephenpfohl @FAccTConference @mikarv @struthious @Borhane_B_H @andrewthesmart heard from a Program chair: "your paper will not be moved to non archival if you have an open ticket and the issue is with ACM, so don’t stress"

Help: #FAccT #FAccT24 @FAccTConference folks, anyone else stuck here with TAPS? If I remember correctly, there used to be a way to see the error log and debug independently…

@davidthewid @FAccTConference @mikarv @struthious @Borhane_B_H @andrewthesmart FWIW, I have been stuck in this state for a few days now

@FAccTConference @mikarv @struthious @Borhane_B_H @andrewthesmart To be clear: I don't see any way to debug this because it doesn't give an error log nor a way to submit other /new files

@PhDemetri Charles Elkan called it in 2001 cseweb.ucsd.edu/~elkan/rescale…

You're working on a model where the outcome is binary. The positive outcome is the minority case, with a prevalence of about 1:9. You decide to upsample the minority case. Your test AUC is 0.78. What happens to your calibration as compared to no resampling?

Excited to share our newest work! 📝 Evaluation of LLMs is hard, especially for health equity. We provide a multifaceted human assessment framework, 7 newly-released adversarial datasets, and perform the largest human eval study on this topic to date. 🧵: arxiv.org/abs/2403.12025