Self-Supervised Visual On-Policy Distillation

  • 类型:arxiv
  • 标识:2608.14144
  • 链接:https://arxiv.org/abs/2608.14144
  • 主分类:multimodal
  • 形态:method
  • 被引:0
  • 被引来源:Semantic Scholar
  • S2被引:0
  • 影响力被引:0
  • TLDR:Self-Supervised Visual On-Policy Distillation (S$^2$VOPD), a simple yet effective method that constructs on-policy learning signals from asymmetric augmented views, systematically explores a broad design space of visual augmentations and uncover that asymmetry matters.
  • 待LLM分类:否
  • 标题中文:自监督视觉 On-Policy Distillation
  • TLDR中文:提出 Self-Supervised Visual On-Policy Distillation(S²VOPD),一种简单有效的方法,通过非对称增强视图构建 on-policy 学习信号,系统地探索了视觉增强的广阔设计空间,并发现非对称性至关重要。
  • 来源文件
  • /inbox/tom/_candidates/2026-08-17-agent-rag-longcontext-candidates.json
  • [S2 enrich]