Paper page - aDSL: Agentic 3D Creation via Joint Agent-Program Design

Source: https://huggingface.co/papers/2608.17975

Abstract

A co-designed domain-specific language and multi-agent system improve LLM-driven 3D program synthesis by using relational operators and iterative execution feedback.

Programmatic representations provide a compelling paradigm for 3D content creation, enabling fine-grained edits, interpretability, and explicit structural control. Yet, agentic workflows that rely on large language models (LLMs) to author 3D programs remain brittle, often failing to translate high-level intent into consistent low-level geometry. We attribute this fragility to a mismatch between existing programmatic interfaces and the reasoning strengths of LLMs, which favor semantic structure and spatial relations over fragile numeric choices. In this paper, we jointly design anAgent-centric Domain-Specific Language(aDSL) and a role-specializedmulti-agent systemto close this gap.aDSLbridges semantic logic and geometric constraints by emphasizing composability and spatial reasoning; it enables agents to manipulate geometry throughrelational operatorsinstead of brittle absolute coordinates. Building onaDSL, our training-freemulti-agent systemfollows aPlan-Execute-Critic loopto decompose requests, synthesize code, and iteratively repair errors and constraint violations using execution feedback. Experiments show that this co-design improves robustness, controllability, and faithfulness to user intent. Our method outperforms prior LLM-based baselines ontext-to-shapeandimage-to-shapetasks while preserving explicit structure, editability, and interpretability. It also enables downstream applications such asarticulated object creationandstructured scene composition. Our code is available at https://github.com/sig-pku/aDSL.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2608\.17975

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.17975 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.17975 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.17975 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.