Vision-language models largely fail to distinguish multimodal puns from adversarial non-puns but gain an average 16.5% F1 improvement from prompt-level and model-level interventions.
is_pun": true/false} IMPORTANT: Output ONLY the JSON object, no additional text or explanation. Note: The biased-to-non-pun variant changes the task description to
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
-
"I See What You Did There": Can Large Vision-Language Models Understand Multimodal Puns?
Vision-language models largely fail to distinguish multimodal puns from adversarial non-puns but gain an average 16.5% F1 improvement from prompt-level and model-level interventions.