LLMs exceed random baseline in CAH humor selection but align modestly with humans and more strongly with each other due to position and content biases.
Respond ONLY with: <number>
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
-
Cards Against LLMs: Benchmarking Humor Alignment in Large Language Models
LLMs exceed random baseline in CAH humor selection but align modestly with humans and more strongly with each other due to position and content biases.