The LLM has very little memory between two tokens of the same message (there might be a little in the sampling algorithm), so it should be equally capable of no

The LLM has very little memory between two tokens of the same message (there might be a little in the sampling algorithm), so it should be equally capable of noticing an error in the same message as a new one.