Paper page - SimWorld Studio: Automatic Environment Generation with Evolving Coding Agent for Embodied Agent Learning

Source: https://huggingface.co/papers/2605.09423

Abstract

SimWorld Studio presents an open-source platform using Unreal Engine 5 and a skill-augmented coding agent to automatically generate evolving 3D environments for embodied agent training through self-evolution and co-evolution mechanisms.

LLM/VLM-based digital agents have advanced rapidly thanks to scalable sandboxes for coding, web navigation, and computer use, which provide rich interactive training grounds. In contrast,embodied agentsstill lack abundant, diverse, and automatically generated3D environmentsfor interactive learning. Existing embodied simulators rely on manually crafted scenes or procedural templates, while recent LLM-based 3D generation systems mainly produce static scenes rather than deployable environments with verifiable tasks and standard learning interfaces. We introduce SimWorld Studio, an open-source platform built onUnreal Engine 5for generating evolving embodied learning environments. At its core isSimCoder, a tool/skill-augmented coding agent that writes and executes engine-level code to construct physically grounded 3D worlds from language/image instructions.SimCoderself-evolves by using verifier feedback (e.g., compilation errors, physics checks, VLM critiques) to revise environments and autonomously add reusable tools and skills to its library. Generated worlds are exported asGym-style environmentsfor embodied agent learning. SimWorld Studio further enablesco-evolutionbetween environment generation and embodied learning:agent performance feedbackguidesSimCoderto generateadaptive curriculanear the learner’s capability frontier, so that environments become increasingly challenging as the embodied agent improves. Three case studies onembodied navigationshow thatself-evolutionimproves generation reliability, generated environments substantially improve embodied agent performance that generalizes to unseen benchmarks, andco-evolutionyields an 18-point success-rate gain over fixed-environment learning and a 40-point gain over an untrained agent.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2605\.09423

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/2605.09423 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.09423 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.09423 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.