1 paper
Seungsu Kim, Jinyoung Choi, Seungmin Baek +1
We present ProgVLA, a compact vision-language-action (VLA) model designed for reliable robot manipulation under tight compute and memory budgets. The model specifically focuses on…