Meta 发布 JEM 框架，自监督学习性能对标 DINOv3
Meta 憋出的自监督新招 JEM，数据量只有 DINOv3 的 1/12 还能赢，做视觉表征的值得看下论文细节。
Meta 研究人员发表论文《Scalable Patch-Level Self-Supervised Learning》，提出学生-教师框架 JEM，通过跨视图对齐 patch 表示并结合显式正则化防止结构坍塌。在受控实验的密集预测任务上，JEM 成绩超过 DINOv2。对比 DINOv3 的官方 checkpoint，JEM 用少 12 倍的数据量仍取得更优表现。
This paper went completely under the radar!
Meta researchers have developed a novel and simple self-supervised learning framework competitive with DINOv2 and DINOv3!
"Scalable Patch-Level Self-Supervised Learning"
The authors introduce JEM, a principled student-teacher method that learns to align patch representations across different views. Additionally, it incorporates explicit regularization to ensure representations are informative and prevent structural collapse.
The framework outperforms DINOv2 in a controlled setting on dense tasks, and even beats the DINOv3 checkpoint despite it being trained on 12x less data.
