one caveat: RWKV-7 can simulate DFA, but it requires feature dimension to be exp(c|S|) where |S| is the number of states of the (reverse-)DFA.
Quote
BlinkDL
@BlinkDL_AI
RWKV-7 "Goose" with Expressive Dynamic State Evolution 
paper is out: huggingface.co/papers/2503.14 RWKV-7 can perform state tracking and recognize all regular languages, while retaining parallelizability of training
paper is out: huggingface.co/papers/2503.14 RWKV-7 can perform state tracking and recognize all regular languages, while retaining parallelizability of training
