Comment by kmchandy

The paper makes a clear claim: "it provides an important and concrete proof: on well-constrained, verifiable reasoning tasks, first-tier performance is no longer the exclusive domain of ultra-large models" And that's exciting.