A Classical Chinese language model internally detects unknown facts via perplexity jumps but fails to express uncertainty in generated text, showing that metacognitive expression does not emerge from language modeling alone.
However, the consistent trend across our tested range (no emergence of metacognitive expression at any scale) suggests this is unlikely
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Internal Knowledge Without External Expression: Probing the Generalization Boundary of a Classical Chinese Language Model
A Classical Chinese language model internally detects unknown facts via perplexity jumps but fails to express uncertainty in generated text, showing that metacognitive expression does not emerge from language modeling alone.