Search papers, labs, and topics across Lattice.
University of Liverpool
1
0
3
0
By generating contrasting inferences about hatefulness, a new adversarial reasoning approach significantly boosts hateful video detection, outperforming existing methods by up to 7% in hate class recall.
