Stephan Rabanser

@steverab

Postdoctoral Researcher @Princeton. Reliable, safe, trustworthy machine learning. Previously: @UofT @VectorInst @TU_Muenchen @Google @awscloud

Princeton, NJ

rabanser.dev

Replies Reposts Media

Pinned
Stephan Rabanser
@steverab
Jul 30
We’ve been hard at work on CRUX 2, the latest iteration of our research stream for open-world evaluations of frontier AI. With growing excitement around recursive self-improvement, we assessed progress toward automated AI R&D using a new method we call shadow evaluations.
2
Stephan Rabanser
@steverab
Sep 15
Highly recommend reading this thoughtful, comprehensive, and well-argued piece from @sayashk and @random_walker on the recent safety incidents and the more general anxiety in our field around loss-of-control: normaltech.ai/p/the-ai-as-no…
Sayash Kapoor
@sayashk
Sep 14
What does it mean to pace the frontier? Over the last month, @random_walker and I have analyzed the loss-of-control incidents at AI companies to understand what technical and policy interventions can improve safety and what companies should do to pace the frontier. The result
Stephan Rabanser
@steverab
Aug 19
Our work on conducting automated AI research with agents has sparked discussion across views on recursive self-improvement. It was also featured in a Nature article where one of the creators of Sakana's AI Scientist describes our harness as "trivially fixable" to boost results.
Why do our findings differ from Sakana’s AI Scientist?
From cruxevals.com
1
Stephan Rabanser
@steverab
Aug 4
Last week we put out a paper on whether AI agents can do open-ended research. Within hours, Google's AI Overview reversed its answer to our main research question from a confident "Yes" to a confident "No." Our arXiv upload was enough to rewrite the authoritative summary. This
Stephan Rabanser
@steverab
Jul 30
We’ve been hard at work on CRUX 2, the latest iteration of our research stream for open-world evaluations of frontier AI. With growing excitement around recursive self-improvement, we assessed progress toward automated AI R&D using a new method we call shadow evaluations.
1
Stephan Rabanser
@steverab
Jul 2
Very excited to discuss the capability-reliability gap in agentic AI at the @farairesearch Seoul Alignment Workshop next week!
FAR.AI
@farairesearch
Jul 2
Confirmed for Seoul Alignment Workshop: @steverab (@Princeton), on the opening panel on the capability-reliability gap. He works on when machine learning systems can be trusted: uncertainty, calibration, knowing when a model should abstain. The premise underneath it: a capable
1