Action chunking in VLAs
mediumsubjective
Practice question
Many vision-language-action models predict a short sequence (chunk) of future actions per inference rather than a single action. Why?
Voice input needs Chrome or Edge. On this browser, type your answer — grading is the same.