I'm no chemist, but from a comp sci perspective the title claim seems overstrong. One can show that certain models of one category outperform certain models of

I'm no chemist, but from a comp sci perspective the title claim seems overstrong. One can show that certain models of one category outperform certain models of another on certain benchmarks, or first demonstrate some impressive result with a mode based on such a contrary insight. But it's very very hard to prove that a whole category - e.g. models eliminating invalid SMILES - won't lead to competitive results.