EESD：编码智能体自纠正的适应性证据自蒸馏

1. 12:00
   arXiv · Software Engineering

   编程智能体的自纠正应承载多少证据？面向自蒸馏的自适应 Dirichlet 证据

   论文提出 Effective-Evidence Self-Distillation（EESD），把执行反馈中的转移支持与证据质量分开建模，用 Dirichlet 后验生成带不确定性惩罚的权重进行 KL 锚定的纠正学习。

Current heat 9·Comparable peak 10（Oct 7）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.