I think that tells you more about the uselessness of SOTA benchmarks.
I think it says more about people's ability to ignore the truth if it doesn't support their world view. Oh you don't want Grok to be SOTA? Then it isn't! Problem solved
alt Hacker News
I think it says more about people's ability to ignore the truth if it doesn't support their world view. Oh you don't want Grok to be SOTA? Then it isn't! Problem solved