An OCR, captioning, sub-label retrieval, and multi-turn VQA pipeline reaches 73.5 percent accuracy and 78.35 AUROC on Facebook Hateful Memes.
In: 2018 International Conference on Frontiers of Information Technology (FIT)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Detecting and Understanding Hateful Contents in Memes Through Captioning and Visual Question-Answering
An OCR, captioning, sub-label retrieval, and multi-turn VQA pipeline reaches 73.5 percent accuracy and 78.35 AUROC on Facebook Hateful Memes.