Normalized Matching Transformer enforces unit-norm embeddings at every Transformer layer and trains with InfoNCE plus hyperspherical uniformity loss, reaching new state-of-the-art accuracy on PascalVOC and SPair-71k while converging faster than prior matching networks.
Deep graph matching consensus
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
RCTEA introduces richness-guided co-training with attention and dual-view consensus to achieve state-of-the-art temporal entity alignment on public benchmarks.
citing papers explorer
-
Normalized Matching Transformer
Normalized Matching Transformer enforces unit-norm embeddings at every Transformer layer and trains with InfoNCE plus hyperspherical uniformity loss, reaching new state-of-the-art accuracy on PascalVOC and SPair-71k while converging faster than prior matching networks.
-
RCTEA: Richness-guided Co-training for Temporal Entity Alignment
RCTEA introduces richness-guided co-training with attention and dual-view consensus to achieve state-of-the-art temporal entity alignment on public benchmarks.