
Daniel Lenton
@DanielLenton1 • 1,533 subscribers
Founder @letsunifyai | YC
Videos

OpenClaw, Hermes Agent and Prime Agent don't meaningfully learn from experience. On Continual-ARC, our new benchmark at unify, none of them were able to memorize anything of value and none were able to generalize to new tasks. Training a small 9B model's weights did both.
Daniel Lenton782,675 次观看 • 11 小时前
没有更多内容可加载