AssemblyBench dataset and AssemblyDyno transformer model enable physics-aware prediction of assembly sequences and trajectories for complex industrial objects from multimodal instructions and 3D shapes.
Manual2Skill++: Connector- aware general robotic assembly from instruction manuals via vision-language models.arXiv preprint arXiv:2510.16344
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
other 1
citation-polarity summary
years
2026 4representative citing papers
ActGaze derives self-supervised visual attention targets from counterfactual image interventions in a vision-language-action policy, improving average success on four precision manipulation tasks from 77.0% to 91.5%.
PART casts 3D part retrieval and assembly as variable-length set prediction with transformers, combining joint segmentation learning with test-time pose refinement.
AssemLM fuses SO(3)-equivariant point-cloud features into a VLM to predict discrete 6D assembly poses, reaching ~89% success on AssemBench and improved real-robot multi-step assembly.
citing papers explorer
-
AssemblyBench: Physics-Aware Assembly of Complex Industrial Objects
AssemblyBench dataset and AssemblyDyno transformer model enable physics-aware prediction of assembly sequences and trajectories for complex industrial objects from multimodal instructions and 3D shapes.
-
ActGaze: Learning Action-Grounded Gaze through Counterfactual Visual Interventions for High-Precision Manipulation
ActGaze derives self-supervised visual attention targets from counterfactual image interventions in a vision-language-action policy, improving average success on four precision manipulation tasks from 77.0% to 91.5%.
-
PART: Learning 3D Part Assembly and Retrieval with Transformers
PART casts 3D part retrieval and assembly as variable-length set prediction with transformers, combining joint segmentation learning with test-time pose refinement.
-
AssemLM: A Spatial Reasoning Multimodal Large Language Model for Robotic Assembly
AssemLM fuses SO(3)-equivariant point-cloud features into a VLM to predict discrete 6D assembly poses, reaching ~89% success on AssemBench and improved real-robot multi-step assembly.