Cortex 2.0: Grounding World Models in Real-World Industrial Deployment

ArXi:2604.20246v1 Announce Type: cross Industrial robotic manipulation demands reliable long-horizon execution across embodiments, tasks, and changing object distributions. While Vision-Language-Action models have nstrated strong generalization, they remain fundamentally reactive. By optimizing the next action given the current observation without evaluating potential futures, they are brittle to the compounding failure modes of long-horizon tasks.