ConsisVLA-4D: Advancing Spatiotemporal Consistency in Efficient 3D-Perception and 4D-Reasoning for Robotic Manipulation

Documenti analoghi

• SemanticVLA: Semantic-Aligned Sparsification and Enhancement for Efficient Robotic Manipulation
  di: Li, Wei, et al.
  Pubblicazione: (2025)
• ST-VLA: Enabling 4D-Aware Spatiotemporal Understanding for General Robot Manipulation
  di: Wu, You, et al.
  Pubblicazione: (2026)
• AnchorVLA4D: an Anchor-Based Spatial-Temporal Vision-Language-Action Model for Robotic Manipulation
  di: Zhu, Juan, et al.
  Pubblicazione: (2026)
• ActiveVLA: Injecting Active Perception into Vision-Language-Action Models for Precise 3D Robotic Manipulation
  di: Liu, Zhenyang, et al.
  Pubblicazione: (2026)
• CogVLA: Cognition-Aligned Vision-Language-Action Model via Instruction-Driven Routing & Sparsification
  di: Li, Wei, et al.
  Pubblicazione: (2025)