Introduces the 'innovation' property of LLMs and proves it is an almost characterization of hallucination while deriving new lower bounds on hallucination rates via missing mass.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
SAI extends the RPNI framework with a splitting operation to identify symbolic automata over monotonic algebras with predicates a <= x < b in the limit, with a proof of polynomial-size characteristic samples.
citing papers explorer
-
Innovation: An Almost Characterization of Hallucination
Introduces the 'innovation' property of LLMs and proves it is an almost characterization of hallucination while deriving new lower bounds on hallucination rates via missing mass.