Published:
Some simple logic games are surprisingly hard for current “Large Reasoning Models”, making this a promising research direction, as recently noted by Andrej Karpathy in response to the TextArena release.
Published:
TL;DR: ML reasoning systems are going through an exciting phase transition that escapes our understanding; we neither can nor want to stop it, but this requires an AI Safety approach similar to autonomous driving (engineering + human accountability) to avoid the pitfalls of AI anthropomorphization.
Published:
After 7 wonderful years leading ML at TRI, it is time for me to embark on a new adventure: I’m thrilled to be joining the early-stage deep tech VC firm Calibrate Ventures as a Partner! Why? Because Calibrate is uniquely positioned to back the most promising startups in Embodied Intelligence.