URMF uses learnable Gaussian posteriors to estimate modality-specific uncertainty and adjust fusion weights for improved multimodal sarcasm detection on MSD and MMSD2 benchmarks.
0: A multi-image benchmark for real-world multimodal sarcasm detection
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
URMF: Uncertainty-aware Robust Multimodal Fusion for Multimodal Sarcasm Detection
URMF uses learnable Gaussian posteriors to estimate modality-specific uncertainty and adjust fusion weights for improved multimodal sarcasm detection on MSD and MMSD2 benchmarks.