Mason Kamb
@MasonKamb
Ph.D. student, Applied Physics @Stanford.
Excited to finally share this work w/ @SuryaGanguli. Tl;dr: we find the first closed-form analytical theory that replicates the outputs of the very simplest diffusion models, with median pixel wise r^2 values of 90%+. arxiv.org/abs/2412.20292
Maybe worth saying, since I think it’s not obvious: this also applies to my job (mathematician)
I'm at ICML presenting this work! Come by on Tuesday to hear about/chat about combinatorial generalization and creativity in diffusion models. x.com/MasonKamb/stat…
Excited to finally share this work w/ @SuryaGanguli. Tl;dr: we find the first closed-form analytical theory that replicates the outputs of the very simplest diffusion models, with median pixel wise r^2 values of 90%+. arxiv.org/abs/2412.20292
Can image-generating AI models really be "creative"? Or is there something deeper at play? My latest for @QuantaMagazine: x.com/QuantaMagazine…
In a recent paper, physicists used two predictable factors to reproduce the “creativity” seen from image-generating AI. @_webbwright reports: quantamagazine.org/researchers-un…
A great @QuantaMagazine article on our theory of creativity in convolutional diffusion models lead by @MasonKamb. See also our paper with new results in version 2: arxiv.org/abs/2412.20292 to be presented as an oral at @icmlconf #icml25 thx @_webbwright !
In a recent paper, physicists used two predictable factors to reproduce the “creativity” seen from image-generating AI. @_webbwright reports: quantamagazine.org/researchers-un…
In a recent paper, physicists used two predictable factors to reproduce the “creativity” seen from image-generating AI. @_webbwright reports: quantamagazine.org/researchers-un…
generational theft won't lead to productive mining or logging, won't make a dent in the deficit, will only render vast swaths of land--including right up the Hood River watershed or the Yosemite Valley foothills--that will never be accessible to the public again
My brother has been digging into this more now. It appears the full plan is coming clearer and shows that its actually planned for a total of 120 million acres… The first phase already exposes the real intent, as most is in and around highly desirable areas of the Tetons,…
Excited to have had this opportunity to talk about the mechanistic origins of diffusion model creativity @KempnerInst last Friday!
NEW: Mason Kamb (@MasonKamb) from @Stanford presents a predictive theory of combinatorial creativity in diffusion models. Watch the video: youtu.be/DP_kGt0-2cg #NeuroAI2025 #AI #ML #NeuroAI
NEW: Mason Kamb (@MasonKamb) from @Stanford presents a predictive theory of combinatorial creativity in diffusion models. Watch the video: youtu.be/DP_kGt0-2cg #NeuroAI2025 #AI #ML #NeuroAI
Looking forward to speaking today at Harvard’s @KempnerInst NeuroAI conference on theories of learning, creativity and reasoning. Looks like a great set of speakers. Virtual attendance is possible: kempnerinstitute.harvard.edu/frontiers-in-n…
This graphic of Trump’s cuts to NASA’s research fleet is insane. Why do this? Who wants this? Who was complaining we were exploring space too much? The only principle seems to be “if it can’t go directly into Trump or Musk’s own pockets, dump it in the trash.”