A 1.2B-parameter vision-language-action policy that chains task progress and the previous plan's unfinished motion across queries reaches 62.8% success on RMBench and 98.8% on LIBERO.
Proceedings of the AAAI Conference on Artificial Intelligence , volume =
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
-
ChainVLA: Chaining Vision-Language-Action Queries through a Unified Execution State for Long-Horizon Manipulation
A 1.2B-parameter vision-language-action policy that chains task progress and the previous plan's unfinished motion across queries reaches 62.8% success on RMBench and 98.8% on LIBERO.