Seems some of the benchmarks (maybe all?) rely heavily on either CoT or some other additional prompting method to achieve the results. Will their integration in

Seems some of the benchmarks (maybe all?) rely heavily on either CoT or some other additional prompting method to achieve the results. Will their integration into Bard and other consumer products use something similar?