Because in science, not admitting an error you noticed is way worse than making one. Trust thrives when people are open and honest about mistakes. Literally nobody is flawless, but the collective system of science works. #Science #Honesty #Trust #Therearenospecificcountriesmentionedinthetweet.Therefore #theresponsewouldbe:
I do worry about public trust in science. But idealized pictures of science can become harmful too if they make honest discussion of uncertainty, mistakes, heuristics, or distributed validation impossible. A picture of science based on perfect individual rigor is not sustainable. #ScienceTrust #UncertaintyInScience #RealityOfScience #Nospecificcountrycodescanbeidentifiedintheprovidedtweet.
To clarify: my point is not that standards or verification do not matter. Quite the opposite. Science works because verification is distributed across a community over time — through reproduction, criticism, use, and extensions — not because scientists never make minor mistakes. #Science #Verification #Community #Nospecificcountrycodescanbeidentifiedinthetweetprovided.
Are you in Prague and do you want to hear about my groups work? Come to say hi next Tuesday to this seminar: https://t.co/wzSO1xlSNl
Congrats to the great #NeurIPS best paper (https://t.co/junOr0NASh) on memorization in diffusion models. But note: producing non-memorized samples is not the same as generalization in the sense of learning structure or rules. Understanding generalization in DMs remains open. -
RT @KempnerInst: Thanks to Lenka Zdeborová (@zdeborova) from @EPFL for a great talk discussing attention-based architectures that operate o… -
