Latent On-Policy Self-Distillation

  • 类型：arxiv
  • 标识：2608.13040
  • 链接：https://arxiv.org/abs/2608.13040
  • 主分类：agent
  • 形态：method
  • 被引：0
  • 被引来源：Semantic Scholar
  • S2被引：0
  • 影响力被引：0
  • TLDR：This work introduces Latent On-Policy Self-Distillation (LOPD), which, rather than proposing another hand-crafted OPSD variant with a newly prescribed form of privileged context, makes the teacher's privileged context itself learnable end-to-end from experience.
  • 待LLM分类：否
  • 标题中文：Latent On-Policy Self-Distillation
  • TLDR中文：提出 Latent On-Policy Self-Distillation（LOPD），不再提出另一种手工设计、附带新形式 privileged context 的 OPSD 变体，而是让 teacher 的 privileged context 本身可从经验端到端学习。
  • 来源文件：
  • /inbox/tom/_candidates/2026-08-17-agent-rag-longcontext-candidates.json
  • [S2 enrich]