Metraipolis › archive › dc_68c2bc016f97
Multimodal Thinking with Renderable Programs — arXiv:2609.30130
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.30130
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T04:50:15.340Z
- SHA-256 of the archived text
fb7e759b85dc97beba8dd132eb211af8c6b30956a4a75a176888c63d041f531a- Stable URL of this page
https://metraipolis.com/documento/dc_68c2bc016f97
How to cite
Multimodal Thinking with Renderable Programs — arXiv:2609.30130 — arXiv (en), https://arxiv.org/abs/2609.30130 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_68c2bc016f97 (Metraipolis archive dc_68c2bc016f97, sha256 fb7e759b85dc97beba8dd132eb211af8c6b30956a4a75a176888c63d041f531a)
BibTeX
@misc{dc68c2bc016f97,
title = {Multimodal Thinking with Renderable Programs — arXiv:2609.30130},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.30130},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_68c2bc016f97; sha256 fb7e759b85dc97beba8dd132eb211af8c6b30956a4a75a176888c63d041f531a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.30130. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Current vision-language models (VLMs) excel at visual content understanding and text-based reasoning, yet their structure limits the advancement of incorporating images into the reasoning chain. Though Omnimodal models have made efforts in unifying text and image generation, they focus on visual tasks in the open-domain, lacking tractability due to rasterized or latent representations of images. We introduce SVGLM, a framework that uses scalable vector graphics (SVG) primitives to connect text and image in reasoning tasks. We exploit the duality of SVG as both image description and text instructions, yielding a more compact, interpretable solution to equip general VLMs with the capability of generating images within the reasoning process. We provide a large curated dataset of SVG-based image editing dataset, as well as the paradigm to tune open-source VLMs. Experiments on a mathematical reasoning benchmark demonstrate that SVGLM achieves strong SVG generation power as well as think-with-image intelligence. Our results highlight SVG as a suitable medium for building more robust digital domain agents, bridging the gap between text-based thinking and pixel-based images. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.CL). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.30130 — 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 2026-10-08T04:27:20.174Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).