Arguably this is already happening: the whole state of the model gets fed through from token to token, and even just shoving a bunch of dashes in between the input tokens and the model's output can improve performance (thinking tokens from the model help a little bit more, but the difference is not as large as you mught expect).
Arguably this is already happening: the whole state of the model gets fed through from token to token, and even just shoving a bunch of dashes in between the input tokens and the model's output can improve performance (thinking tokens from the model help a little bit more, but the difference is not as large as you mught expect).