That would be a ton of problems for a small team of PhD/Grad level experts to solve (for GPQA Diamond, etc) in a short time. Remember, on EpochAl Frontier Math,

That would be a ton of problems for a small team of PhD/Grad level experts to solve (for GPQA Diamond, etc) in a short time. Remember, on EpochAl Frontier Math, these problems require hours to days worth of reasoning by humans The author also suggested this is a new architecture that uses existing methods, like a Monte Carlo tree search that deepmind is investigating (they use this method for AlphaZero) I don't see the point of colluding for this sort of fraud, as these methods like tree search and pruning already exist. And other labs could genuinely produce these results