A local cross-layer routing mechanism for VLA models, feeding each action-decoder block features from a small window of adjacent VLM layers, improves success rates on LIBERO, CALVIN, and LIBERO-Plus.
Proceedings of The 8th Conference on Robot Learning, (CoRL) , series=
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
-
LIRA: Local Cross-Layer Information Routing for Vision-Language-Action Decoding
A local cross-layer routing mechanism for VLA models, feeding each action-decoder block features from a small window of adjacent VLM layers, improves success rates on LIBERO, CALVIN, and LIBERO-Plus.