

Can We Trust the Grader? Drinks on Evals and Expert Judgment
Registration
Approval Required
Your registration is subject to host approval.
About Event
We're seeing models beat benchmarks, and AI judges are increasingly standing in for human review. But when the stakes are real, and high, like in medicine, law, and finance, who decides what's right?
Join the GLG Frontier team to discuss what good evaluation looks like once models outgrow the tests, and where human expertise and judgment matters most.
Swing by Der Biergarten after your NeurIPS sessions for a beer and a low-key chat with our team and other researchers working on evals and post-training.
📍 Park Bar, 150 Walton St NW (about an 8-minute walk from GWCC)
🕔 5:00–7:00 PM ET
🍻 Drinks on GLG Frontier