Search papers, labs, and topics across Lattice.
1
0
3
2
By explicitly modeling multi-view visual-semantic relationships, MViR achieves state-of-the-art fake news detection, suggesting that subtle image-text alignments are critical for identifying misinformation.
