Chi Nguyen

How do LLMs reason about playing games against copies of themselves? 🪞We made the first LLM decision theory benchmark to find out. 🧵1/10

How close are current AI agents to automating AI R&D? Our new ML research engineering benchmark (RE-Bench) addresses this question by directly comparing frontier models such as Claude 3.5 Sonnet and o1-preview with 50+ human experts on 7 challenging research engineering tasks.

@FinishItPod It happened! I was late but you were early. Thank you @FinishItPod !

@FinishItPod After 6 weeks of listening, I've just finished episode 106 of (yes, I've listened to the first 9 episodes in one day) Sure hope you still do complis and concris because I want mine in 4 months... Also, the the abominable snowman 1 isn't on spotify

@mlipsitch China: second. attack rate child vs. adult (~12% both) thelancet.com/journals/lanin… Ch: Comp. infection rates w contact patterns (child. less suceptible, riskier behavior) science.sciencemag.org/content/early/… Ger: Indir. evidence infectiousness: viral load (~same all ages) bit.ly/2XR2x4B

#Covidtwitter Is there an analysis out there of probability of infection among known contacts of a case according to source age (besides RIVM) and according to contact age?