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