Viewing profile — arb_

I think they tweet prototypes and then don't ship most prototypes. Seems reasonable to me, otherwise you have instant bloat.

[dead]

As models improve, human preference will become worse as a proxy measurement (e.g. as model capabilities surpass the human's ability to judge correctness at a glance). This can be …

Empirically, they have reduced hallucinations. Where do OpenAI / Anthropic claim that their models won't hallucinate?