The Compression Gap: Why Discrete Tokenization Limits Vision-Language-Action Model Scaling
返回详情
VLA / Vision-Language-Action 每日论文卡