arXivYijiang Li, Bingyang Wang, Yijun Liang, Yunjie Tian, Di Fu, Nuno VasconcelosThu, Aug 6, 2026, 10:18 AM PDT
score 14.7
AI models now train themselves without human or bigger-model help
Original: On-Policy Self-Distillation without Any Supervision
Source: arxiv.org ↗
Writing ELI5 summary…