RAMA, a new benchmark with defective instructions, and RationalVLA, a dual-system model with <ACT> and <REJ> tokens, let a robot reject infeasible commands while still performing unseen executable tasks, with higher success and longer task sequences than baselines.
Vision-language foundation models as effective robot imitators,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
RationalVLA: A Rational Vision-Language-Action Model with Dual System
RAMA, a new benchmark with defective instructions, and RationalVLA, a dual-system model with <ACT> and <REJ> tokens, let a robot reject infeasible commands while still performing unseen executable tasks, with higher success and longer task sequences than baselines.