Aivora

#autoregressive

Autoregressive

1 article

Long-WAM: Scaling Context for Real-Time World-Action Models
arXivRobotics

Long-WAM: Scaling Context for Real-Time World-Action Models

Long-WAM:突破即時控制延遲,擴展機器人世界動作模型的長上下文

Long-WAM scales the visual history context of world-action models under real-time constraints, demonstrating that autoregressive pretraining is key to unlocking the power of long video memories.

2 min read