Alexandra Olteanu
@aolteanu.bsky.social
Ethical/Responsible AI. Rigor in AI. Opinions my own. Principal Researcher @ Microsoft Research. Grumpy eastern european in north america. Lovingly nitpicky.
Interested in any of these areas? Apply here:
apply.careers.microsoft.com/careers/job/...
apply.careers.microsoft.com/careers/job/...
December 5, 2025 at 8:52 PM
Interested in any of these areas? Apply here:
apply.careers.microsoft.com/careers/job/...
apply.careers.microsoft.com/careers/job/...
If you are interested in conceptual clarity, construct validity, benchmarks, and/or anthropomorphic AI behavior, design, or impacts you should consider applying! 🤩 If you are interested in consent and ownership frameworks for AI artifacts, I would really love to talk to you! 😍
December 5, 2025 at 8:52 PM
If you are interested in conceptual clarity, construct validity, benchmarks, and/or anthropomorphic AI behavior, design, or impacts you should consider applying! 🤩 If you are interested in consent and ownership frameworks for AI artifacts, I would really love to talk to you! 😍
Reposted by Alexandra Olteanu
Our forthcoming NeurIPS position paper, led by @aolteanu.bsky.social, makes this argument (along with several related ones) in more depth. Rigorous AI/ML work should flow from explicit and rigorous premises, not just have a final evaluation that checks some rigor boxes. arxiv.org/abs/2506.14652
Rigor in AI: Doing Rigorous AI Work Requires a Broader, Responsible AI-Informed Conception of Rigor
In AI research and practice, rigor remains largely understood in terms of methodological rigor -- such as whether mathematical, statistical, or computational methods are correctly applied. We argue th...
arxiv.org
November 7, 2025 at 12:19 PM
Our forthcoming NeurIPS position paper, led by @aolteanu.bsky.social, makes this argument (along with several related ones) in more depth. Rigorous AI/ML work should flow from explicit and rigorous premises, not just have a final evaluation that checks some rigor boxes. arxiv.org/abs/2506.14652
Perhaps not as much about how real is or is not, but this is a paper that substantially shaped my views on this topic (I have also been surprised at times about how different folks' conceptualizations of reproducibility can be) cs.uwaterloo.ca/~brecht/cour...
cs.uwaterloo.ca
September 30, 2025 at 6:21 PM
Perhaps not as much about how real is or is not, but this is a paper that substantially shaped my views on this topic (I have also been surprised at times about how different folks' conceptualizations of reproducibility can be) cs.uwaterloo.ca/~brecht/cour...
As we prepare the camera-ready version of this paper, I am also reflecting on how to make this work handier and more useful: rigor cards to make the different facets of rigor easier to grasp? workshops to provide a forum for discussion and debates? something else that would be helpful to you?
September 29, 2025 at 11:13 PM
As we prepare the camera-ready version of this paper, I am also reflecting on how to make this work handier and more useful: rigor cards to make the different facets of rigor easier to grasp? workshops to provide a forum for discussion and debates? something else that would be helpful to you?
Not sure if it has what you need or if they are still collecting but this might be worth checking out archive.org/details/twit...
July 31, 2025 at 2:41 PM
Not sure if it has what you need or if they are still collecting but this might be worth checking out archive.org/details/twit...
tiny but perhaps in their defense Cailler is probably the best chocolate in the world 🍫
July 7, 2025 at 11:52 PM
tiny but perhaps in their defense Cailler is probably the best chocolate in the world 🍫