A paired-frame training objective makes cloud VLA features delay-tolerant, letting a lightweight edge head keep acting with stale remote advice and fresh local vision.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Latency-Tolerant Cloud-Edge Collaborative Vision-Language-Action Models via Emergent Representational Specialization
A paired-frame training objective makes cloud VLA features delay-tolerant, letting a lightweight edge head keep acting with stale remote advice and fresh local vision.