Metraipolis › archive › dc_cab435679c40

ToolArtist: Tool-Using Unified Multimodal Models for Agentic Image Generation — arXiv:2608.04436

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.04436
License (as archived)
CC0-1.0
Archived
2026-10-31T11:47:54.547Z
SHA-256 of the archived text
963b8f293db8fdf9683c40e44182eba9d3dbe9e458510dbb75423de71adaaa59
Stable URL of this page
https://metraipolis.com/documento/dc_cab435679c40

How to cite

ToolArtist: Tool-Using Unified Multimodal Models for Agentic Image Generation — arXiv:2608.04436 — arXiv (en), https://arxiv.org/abs/2608.04436 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cab435679c40 (Metraipolis archive dc_cab435679c40, sha256 963b8f293db8fdf9683c40e44182eba9d3dbe9e458510dbb75423de71adaaa59)
BibTeX
@misc{dccab435679c40,
  title = {ToolArtist: Tool-Using Unified Multimodal Models for Agentic Image Generation — arXiv:2608.04436},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.04436},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cab435679c40; sha256 963b8f293db8fdf9683c40e44182eba9d3dbe9e458510dbb75423de71adaaa59},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.04436. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Text-to-image (T2I) models can produce visually compelling images, yet they remain limited on open-world tasks that require complex semantic understanding, multi-step reasoning, and the integration of external world knowledge. Existing efforts introduce agent capabilities into image generation, but they either prescribe a fixed workflow or place only a subset of the open-world image generation process under agent control. Consequently, reasoning, tool invocation, and image generation are not coordinated by a single policy. We propose ToolArtist, a fully agentic image generation model obtained by post-training a Unified Multimodal Model (UMM). ToolArtist dynamically orchestrates reasoning, external tool use, and native image generation within one unified policy. During Supervised Fine-Tuning (SFT), we equip a teacher agent with search tools alongside an image-generation tool. We then convert the collected trajectories into a UMM compatible format, where the image-generation tool is concealed while the resulting generated images are retained. During Reinforcement Learning (RL), we develop an agentic RL infrastructure for UMMs and introduce Reason-Act-Draw GRPO (RAD-GRPO), which uses complementary intent and quality rewards to jointly optimize the model. Experiments show that placing the entire open-world image-generation process under an agent policy consistently outperforms approaches with fixed pipelines or only partially agent-controlled components. We release the training data and the complete post-training infrastructure.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-06 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.04436 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-31T11:37:40.671Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-06?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).