A new model, MICL, learns sarcasm from contradictions across words, objects, and emotions, and reports top accuracy on the MMSD benchmark and on a new spurious-correlation test set.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multi-View Incongruity Learning for Multimodal Sarcasm Detection
A new model, MICL, learns sarcasm from contradictions across words, objects, and emotions, and reports top accuracy on the MMSD benchmark and on a new spurious-correlation test set.