Contrastive Learning or Masked Autoencoder? Understanding and Improving Self-Supervised Knowledge Distillation

Lying at the intersection of self-supervised learning (SSL) and knowledge distillation (KD), Self-supervised KD (SSKD) differs from classical KD frameworks by assuming the teacher model is pretrained without labels. As SSKD relies on SSL-pretrained teachers, recent works have largely employed masked...

Fuld beskrivelse

Bibliografiske detaljer

Principais autores:	Taegoo Kang, Sung-Ho Bae, Chaoning Zhang
Format:	Artigo
Sprog:	Inglês
Udgivet:	IEEE 2026-01-01
Serier:	IEEE Access
Fag:	Deep learning knowledge distillation self-supervised learning masked image modeling contrastive learning
Online adgang:	https://ieeexplore.ieee.org/document/11417782/
Tags:	Tilføj Tag Ingen Tags, Vær først til at tagge denne postø!