Skip to content
Hot eventLive

TMT:VLA策略运行时后门检测

1 reports1 sources3 hr ago updated

Get the story

AI overview

2026年10月8日,arXiv Robotics(一手)报道提出TMT,一种面向视觉-语言-动作(VLA)策略的运行时后门检测器。该方法基于Token Manifold与潜在Transition建模,用于在触发器出现时识别VLA策略产生的恶意动作,适用于未见任务场景。目前报道仅给出方法概述,未披露实验数据或评估结果。

Generated from reports · updated 3 hr ago

Timeline

Follow the coverage from different angles.

Oct 8, 2026
  1. arXiv · Robotics
    TMT:面向未见任务下视觉-语言-动作策略的运行时后门检测

    提出 TMT,一种基于 Token Manifold 与潜在 Transition 建模的运行时后门检测器,用于发现视觉-语言-动作(VLA)策略在触发器出现时的恶意动作。

Heat trend

Current heat 9·Comparable peak 10(Oct 8)·Comparable change over 24 hours –

02.557.510Oct8Oct8Oct8Oct8

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.