Open-access LLMs detect intrinsic hallucinations in paraphrase and translation with moderate success, but small NLI models match or beat them on several language pairs.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Can LLMs Detect Intrinsic Hallucinations in Paraphrasing and Machine Translation?
Open-access LLMs detect intrinsic hallucinations in paraphrase and translation with moderate success, but small NLI models match or beat them on several language pairs.