SOURCE-LINKED INTELLIGENCE

VICAL: Vicinal Consistency Alignment for Long-Tailed Visual Recognition

arXiv · AI, language, vision and robotics · article · Sep 4, 2026 · UTC

Multi-expert models have become the dominant paradigm for long-tailed learning, largely attributed to their presumed ability to benefit from expert diversity. However, we revisit this central assumption and reveal that diversity induced by logit adjustment or explicit regularizers does not guarantee better ensemble accuracy. Our work suggests that multi-expert models benefit more from variance reduction than diversity maximization. We introduce \textbf{VICAL}, a \textbf{VI}cinal \textbf{C}onsistency \textbf{AL}ignment framework that improves long-tailed recognition not by enforcing expert dive

Read original source ↗ Open in workspace

recordType

paper

region

Global