Evaluation is All You Need: Strategic Overclaiming of LLM Reasoning Capabilities Through Evaluation Design.

Lin Sun 0010, Weihong Lin, Jinzhu Wu, Yongfu Zhu 0001, Xiaoqi Jian, Guangxiang Zhao, Change Jia, Linglin Zhang, Saier Hu, Yuhan Wu 0001, Xiangzheng Zhang

21 Jan 2026CoRR 2025EveryoneCC BY-SA 4.0
Loading