1
Confidence Geometry Reveals Trace-Level Correctness in Large Language Model Reasoning
置信度几何方法揭示大语言模型推理中痕迹级正确性的全新框架，附开源代码。
arXiv:2605.16824v1 Announce Type: new Abstract: Large language models (LLMs) generate not only reasoning text, but also token-level confidence traject…