Put five frontier models in one thread. They compare answers and catch each...
https://trentonsexcellentthoughtss.evergrovio.com/posts/what-does-18-of-23-fact-checks-where-the-correcting-model-was-right-mean-2
Put five frontier models in one thread. They compare answers and catch each other’s mistakes, so you can research and decide with more confidence, without switching tabs