The ScaleVLA paper combines a 552B DeepSeek-V4.1-Flash VLM with a trainable action DiT to evaluate whether a larger vision-language backbone can reduce the demonstration requirements of robot policies.
Source:explainx.ai · Read Original
The ScaleVLA paper combines a 552B DeepSeek-V4.1-Flash VLM with a trainable action DiT to evaluate whether a larger vision-language backbone can reduce the demonstration requirements of robot policies.
Source:explainx.ai · Read Original
Leave a Reply