微软When2Think:让大模型学会难题多想、简单题少想
When2Think: Learning Difficulty-Aware Length Control for Efficient Hybrid Reasoning Models 作者:Jaejun Shim, HyunJin Kim, Young Jin Kim, JinYeong Bak 核心发表机构:Sungkyunkwan University、Microsoft 论文链接:arXiv:2609.19671…
2026/9/28 19:57:03