SLLCP：用保形预测校准VLM的自动驾驶安全监测

1. 12:00
   arXiv · Robotics

   SLLCP：用 VLM 做自动驾驶局部共形安全监测

   研究者提出 Split Label-Localized Conformal Prediction（SLLCP），在冻结的 VLM 之上加后处理校准层，把不可靠预测转成概率校准的安全预测集，并按场景局部加权历史经验计算不确定性阈值。

Current heat 6·Comparable peak 10（Oct 5）·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.