#artificial intelligence
May 2026
Internalizing Curriculum Judgment for LLM Reinforcement Fine-Tuning
METIS (METacognitive Internalized Self-judgment), a novel framework that internalizes curriculum judgment as a native capability, establishes a simple, closed-loop, and highly efficient curriculum internalization paradigm for LLM reinforcement fine-tuning.
Hang Zheng, Yi-Ning Ma, K. Gunasekaran et al.
· arXiv.org · 0 citations