MV-WAM reports 55.7% simulation and 77.5% real-world success rates by aligning heterogeneous visual and action manifolds through causal masking and value-guided rollback.
Learning beyond euclid: Curvature-adaptive generalization for neural net- works on manifolds.arXiv preprint arXiv:2507.02999, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
MV-WAM: Manifold-Aware World Action Model with Value Augmentation
MV-WAM reports 55.7% simulation and 77.5% real-world success rates by aligning heterogeneous visual and action manifolds through causal masking and value-guided rollback.