analiza aiqbench
BrickAnything advances generative modeling by integrating physical constraints—such as structural stability and discrete assembly rules—directly into the autoregressive generation of 3D brick structures. By employing structure-aware tree tokenization to model local attachment relations, this framework addresses a critical gap where previous methods failed when target shapes lacked feasible geometric solutions under predefined buildability limits. The approach demonstrates that aligning sequence prediction with physical construction processes significantly improves validity and reduces regeneration rates compared to heuristic optimization strategies.
BrickAnything: Geometry-Conditioned Buildable Brick Generation with Structure-Aware Tokenization
arXiv:2605.26182v2 Announce Type: replace Abstract: Generating physically buildable brick structures from 3D shapes requires more than geometric reconstruction: the output must also satisfy discrete part constraints and structural stability. Existing brick generation methods either rely on heuristic optimization, which can break down when the target 3D shape does not admit a feasible structure under predefined constraints, or generate brick sequences without explicitly modeling the underlying 3D geometry and assembly relations. In this work, we present BrickAnything, a geometry-conditioned autoregressive framework for generating buildable brick structures from diverse 3D representations. BrickAnything uses point clouds as a unified geometric interface and predicts brick sequences that reconstruct the target shape under assembly constraints. To model structural dependencies among bricks, we introduce a structure-aware tree tokenization, which represents brick structures through local attachment relations. This formulation makes sequence generation more consistent with the physical construction process, and reduces invalid intermediate states. We further introduce preference-based alignment post-training, validity-constrained decoding and adaptive rollback to improve buildability objectives such as stability and geometric fidelity. Extensive experiments demonstrate that BrickAnything produces geometrically faithful and physically realizable brick structures, and that the proposed tokenization effectively reduces rollback and regeneration compared with conventional ordering strategies.
Abstract:Generating physically buildable brick structures from 3D shapes requires more than geometric reconstruction: the output must also satisfy discrete part constraints and structural stability. Existing brick generation methods either rely on heuristic optimization, which can break down when the target 3D shape does not admit a feasible structure under predefined constraints, or generate brick sequences without explicitly modeling the underlying 3D geometry and assembly relations. In this work, we present BrickAnything, a geometry-conditioned autoregressive framework for generating buildable brick structures from diverse 3D representations. BrickAnything uses point clouds as a unified geometric interface and predicts brick sequences that reconstruct the target shape under assembly constraints. To model structural dependencies among bricks, we introduce a structure-aware tree tokenization, which represents brick structures through local attachment relations. This formulation makes sequence generation more consistent with the physical construction process, and reduces invalid intermediate states. We further introduce preference-based alignment post-training, validity-constrained decoding and adaptive rollback to improve buildability objectives such as stability and geometric fidelity. Extensive experiments demonstrate that BrickAnything produces geometrically faithful and physically realizable brick structures, and that the proposed tokenization effectively reduces rollback and regeneration compared with conventional ordering strategies.
Artificial Intelligence (cs.AI); Graphics (cs.GR)
arXiv:2605.26182 [cs.AI]
(or arXiv:2605.26182v2 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2605.26182 Focus to learn more arXiv-issued DOI via DataCite
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.
Text integral sub licența deschisă a publicației · 285 cuvinte
Subiecte conexe
Direct Preference Optimization Beyond Chatbots
Hugging Face · 3 iun. 2026
TreeRef-BFN: Equivariance-Free De Novo Molecule Generation based on 2D Topology and Internal 3D Geometry
arXiv · 30 sept. 2026
The Entangled Dimensions of Decision Theory
LessWrong · 30 iul. 2026
Physics-Guided Generative AI for Property-Targeted 3D Porous Media Design
arXiv · 28 iul. 2026