Are our machine learning force fields too accurate? So fast is the development of this field and so slow the traditional publishing process that I can only hope that these biased opinions, disguised as a review, are still relevant: doi.org/10.1063/5.02...

Nimble work from @jasonmast.bsky.social and colleagues covering the latest attempt by the Administration to strangle biomedical research in the United States: “This indirect cost cut, were it permitted to stand, will result in our children and grandchildren living shorter and less healthy lives"