2 papers
cs.RO2026
MINERVA: How Small Can a Manipulation Policy Be and Still Solve LIBERO?
Kohei Sendai, Tatsuya Matsushima, Yusuke Iwasawa
Vision-language-action (VLA) models with billions of parameters now dominate the LIBERO manipulation benchmark, but the model capacity actually required by the benchmark remains un…
cs.RO2025
Leave No Observation Behind: Real-time Correction for VLA Action Chunks
Kohei Sendai, Maxime Alvarez, Tatsuya Matsushima +2
To improve efficiency and temporal coherence, Vision-Language-Action (VLA) models often predict action chunks; however, this action chunking harms reactivity under inference delay…