Metraipolis › archive › dc_4fc693149bd2

From Corpora to Co-Evolving Capabilities: Capability-Centric Data Design for Generalist Image Generation — arXiv:2608.18076

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.18076
License (as archived): CC0-1.0
Archived: 2026-10-18T14:33:52.114Z
SHA-256 of the archived text: 332edeab39244b03546aefebbc8e4d968c4b95293f3f1233648c0800ba805e49
Stable URL of this page: https://metraipolis.com/documento/dc_4fc693149bd2

How to cite

From Corpora to Co-Evolving Capabilities: Capability-Centric Data Design for Generalist Image Generation — arXiv:2608.18076 — arXiv (en), https://arxiv.org/abs/2608.18076 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4fc693149bd2 (Metraipolis archive dc_4fc693149bd2, sha256 332edeab39244b03546aefebbc8e4d968c4b95293f3f1233648c0800ba805e49)

BibTeX

@misc{dc4fc693149bd2,
  title = {From Corpora to Co-Evolving Capabilities: Capability-Centric Data Design for Generalist Image Generation — arXiv:2608.18076},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.18076},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4fc693149bd2; sha256 332edeab39244b03546aefebbc8e4d968c4b95293f3f1233648c0800ba805e49},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.18076. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large-scale image generation has benefited from advances in data scale, quality, rebalancing, and recaptioning, yet conventional pipelines typically optimize task-specific datasets in isolation. A central challenge is not only how to curate each task-specific corpus, but also how to organize heterogeneous supervision according to the dependencies among generative capabilities. We present a \textbf{capability-driven data infrastructure} that couples capability-specific supervision construction with capability-aligned curriculum scheduling. Its three specialized yet interoperable data engines build complementary relational supervision for text-image grounding, inter-image transformation, and image-knowledge association, while caption experts align T2I and editing supervision across tasks and granularities. A multi-stage curriculum jointly evolves task composition, visual-concept distribution, data quality, and image resolution along the dependency order of capability acquisition, with capability-aware evaluation closing the loop through targeted retrieval, expert construction, and gap-aware resampling. At scale, the framework curates a 440M-image T2I corpus, 120M editing pairs, and over 27M image-entity pairs. With this infrastructure, we train multimodal diffusion models at two scales from scratch, with 3B and 6B sizes respectively. We conduct quantitative evaluation on CPI-Bench, along with qualitative evaluations across diverse text-to-image and editing scenarios. Experimental results present broad visual coverage, versatile rendering, and effective transfer across generative capabilities.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-19 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.18076 — 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 2026-10-18T14:27:54.483Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-19?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).