Visual manuals communicate procedural knowledge through stepwise diagrams, but translating these instructions into reliable long-horizon robotic execution remains challenging. We introduce JARVIS, a method-agnostic benchmark for autonomous visual-manual-guided execution without additional language instructions. JARVIS...
M-Diffusion Planner is proposed with a dual-decoupling design: separating intent understanding from trajectory planning, and general planning capability learning from strategy alignment, to mitigate degradation of pretrained planning capabilities.
The key to the approach is a discrepancy-aware dynamic fusion module, which treats cross-modality inconsistency as an explicit indicator of deception, and creates an interaction representation from which a sample-specific reliability level of each modality can be inferred and dynamically adjusted.
Yuanya Zhuo, V. M. Baskaran, Lillian-Yee-Kiaw Wang et al.· Journal of King Saud Univers...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.