Stochastic Deep Supervision Proposed for Early-Exit Face Recognition in Vision Transformers
A new arXiv paper introduces stochastic deep supervision for gallery-consistent early-exit face recognition, addressing the fixed-depth inference limits of Vision Transformers in resource-constrained settings.
Read the full story at arXiv cs.CV ↗
Timeline · 1 report
- 2026-11-10 05:00 · arXiv cs.CV
Stochastic Deep Supervision for Gallery-Consistent Early-Exit Face Recognition