aiqbench analysis
This research introduces AnchorPose, a novel framework using Bayesian Flow Networks to generate geometrically consistent poses for metal-organic frameworks (MOFs) by explicitly modeling how rotation errors interact with block geometry. By integrating local structural constraints into the generative process, the method improves assembly accuracy over traditional all-atom or block-level baselines, addressing a key challenge in materials design via machine learning.
AnchorPose for Geometry-Aware MOF Assembly through Meso-Grained Pose Generation
arXiv:2610.05843v1 Announce Type: cross Abstract: Predicting metal-organic framework (MOF) structures from given building blocks requires recovering their positions and orientations in a periodic crystal. The spatial effects of rotation errors are geometry-dependent and anisotropic. The same angular error can produce different atomic displacements depending on block size, shape, and rotation axis. Angular error alone, without reference to the specific block geometry, therefore cannot fully describe the spatial consequences of a pose error. We introduce AnchorPose, a meso-grained pose generation framework that incorporates this geometric dependence into its generative representation. It represents each block through a small set of representative atoms, combines their local geometry with the current spatial state, and generates their coordinates with Bayesian Flow Networks. Known atom correspondences enable rigid alignment to recover complete building-block poses and return geometrically consistent points to the generation process. This design connects point-level spatial prediction with block-level structural constraints. Geometry participates in the pose state and its prediction, while rigid reconstruction preserves intra-block structure without treating all atomic coordinates as assembly variables. On the MOF benchmark, AnchorPose improves single-candidate match rates over the compared block-level and all-atom baselines.
Abstract:Predicting metal-organic framework (MOF) structures from given building blocks requires recovering their positions and orientations in a periodic crystal. The spatial effects of rotation errors are geometry-dependent and anisotropic. The same angular error can produce different atomic displacements depending on block size, shape, and rotation axis. Angular error alone, without reference to the specific block geometry, therefore cannot fully describe the spatial consequences of a pose error. We introduce AnchorPose, a meso-grained pose generation framework that incorporates this geometric dependence into its generative representation. It represents each block through a small set of representative atoms, combines their local geometry with the current spatial state, and generates their coordinates with Bayesian Flow Networks. Known atom correspondences enable rigid alignment to recover complete building-block poses and return geometrically consistent points to the generation process. This design connects point-level spatial prediction with block-level structural constraints. Geometry participates in the pose state and its prediction, while rigid reconstruction preserves intra-block structure without treating all atomic coordinates as assembly variables. On the MOF benchmark, AnchorPose improves single-candidate match rates over the compared block-level and all-atom baselines.
Computational Engineering, Finance, and Science (cs.CE); Materials Science (cond-mat.mtrl-sci); Machine Learning (cs.LG)
arXiv:2610.05843 [cs.CE]
(or arXiv:2610.05843v1 [cs.CE] for this version)
https://doi.org/10.48550/arXiv.2610.05843 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 284 words
Related coverage
AssemLM: A Spatial Reasoning Multimodal Large Language Model for Robotic Assembly
arXiv · Jun 12, 2026
Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering
arXiv · Oct 1, 2026
VGM-VS: Rethinking Visual Geometry Model for High-Precision Visual Servoing
arXiv · Sep 24, 2026
LLM2R: From Large Language Models to 3D Reassemblers
arXiv · Oct 15, 2026