arXivJiazhen Huang, Xiao Chen, Xiao Luo, Yong Dai, Senkang Hu, Yuzhi ZhaoWed, May 27, 2026, 10:49 AM PDT
score 16.5
LLM learns better math by validating reusable skill fragments
Original: Skill-Conditioned Gated Self-Distillation for LLM Reasoning
Source: arxiv.org ↗
Writing ELI5 summary…