Predicate frequency is systematically biased in NLI entailment data, fine-tuned LLMs increasingly rely on this frequency cue, and the cue correlates with WordNet hypernym frequency.
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
-
LLMs are Frequency Pattern Learners in Natural Language Inference
Predicate frequency is systematically biased in NLI entailment data, fine-tuned LLMs increasingly rely on this frequency cue, and the cue correlates with WordNet hypernym frequency.