arXivDachuan Shi, Hanlin Zhu, Xiangchi Yuan, Wanjia Zhao, Kejing Xia, Wen Xiao, Wenke LeeTue, May 19, 2026, 9:28 AM PDT
score 16.5
AI model checks draft answers before deep reasoning
Original: CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic Reasoning
Source: arxiv.org ↗
Writing ELI5 summary…