RoboTTT scales robot policy context to 8,000 timesteps using test-time training

Stanford SVL and NVIDIA Robotics introduced RoboTTT, which uses test-time training to give robot policies up to 8,000 timesteps of context at constant inference cost.

The source reports that 8K-context pretraining beats 1K by 62%, and that performance keeps improving from 128 to 8K timesteps with no sign of saturation. The authors also describe one-shot imitation from human video and in-episode error recovery.