“Liquid” neural nets, based on a worm’s nervous system, can transform their underlying algorithms on the fly, giving them unprecedented speed and adaptability.

The titile is a bit misleading since liquid networks are not new. A core (and very interesting) point of the article is that they used to be relatively slow (and impractical) when calculating the equations representing “synapses”. It seems recent advancements have discovered a closed form approximation of this that makes them much more performant.