I don’t imagine a model capable of the first part would require being told not to assume a decipherment is correct
Humans capable of the first part regularly make the mistake of assuming that their decipherment is correct even if it's not internally consistent and fails to adequately explain most of the data.
alt Hacker News
Humans capable of the first part regularly make the mistake of assuming that their decipherment is correct even if it's not internally consistent and fails to adequately explain most of the data.