AutoRegressive Generation with B-rep Holistic Token Sequence Representation

2026-01-23

AI Summary

This work proposes a novel approach to boundary representation (B-rep) generation by unifying geometric and topological information into a single token sequence, overcoming the incompatibility of conventional graph-based representations with sequential generative models. The method introduces a structured sequence representation composed of three hierarchical token types—geometry, position, and face index—and employs a causal-masked, decoder-only Transformer for autoregressive modeling. By moving beyond traditional decoupled representations, this framework enables end-to-end sequential generation of B-reps. Experimental results demonstrate state-of-the-art performance on B-rep generation tasks, validating the effectiveness and feasibility of the holistic token sequence formulation.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Previous representation and generation approaches for the B-rep relied on graph-based representations that disentangle geometric and topological features through decoupled computational pipelines, thereby precluding the application of sequence-based generative frameworks, such as transformer architectures that have demonstrated remarkable performance. In this paper, we propose BrepARG, the first attempt to encode B-rep's geometry and topology into a holistic token sequence representation, enabling sequence-based B-rep generation with an autoregressive architecture. Specifically, BrepARG encodes B-rep into 3 types of tokens: geometry and position tokens representing geometric features, and face index tokens representing topology. Then the holistic token sequence is constructed hierarchically, starting with constructing the geometry blocks (i.e., faces and edges) using the above tokens, followed by geometry block sequencing. Finally, we assemble the holistic sequence representation for the entire B-rep. We also construct a transformer-based autoregressive model that learns the distribution over holistic token sequences via next-token prediction, using a multi-layer decoder-only architecture with causal masking. Experiments demonstrate that BrepARG achieves state-of-the-art (SOTA) performance. BrepARG validates the feasibility of representing B-rep as holistic token sequences, opening new directions for B-rep generation.

Problem

• B-rep
• sequence-based generation
• geometric and topological representation
• autoregressive modeling
• holistic token sequence

Innovation

• B-rep
• holistic token sequence
• autoregressive generation
• transformer architecture
• geometric-topological encoding