Showing cs.ROShow all
2 papers · 1 filter
cs.RO2026
LIBERO-Dynamic: Benchmarking Robotic Manipulation in Dynamic Environments
Shaojie Ren, Jingwen Sun, Zezhi Liu +4
Most simulation benchmarks for robotic manipulation keep target objects static and treat policy inference as instantaneous. These assumptions simplify evaluation but hide how a mov…
cs.RO2026
VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Model
Jingwen Sun, Wenyao Zhang, Zekun Qi +6
Pretraining Vision-Language-Action (VLA) policies on internet-scale video is appealing, yet current latent-action objectives often learn the wrong thing: they remain anchored to pi…