Metraipolis › archive › dc_2f38dc9b89eb
DEPICT: Scoring Text-to-Image Alignment by Answer Agreement — arXiv:2610.03617
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.03617
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T09:08:54.210Z
- SHA-256 of the archived text
6b339376c1e4fa565f58dd81b4aad0777ec413eb14653fbda6cb24bd46e66583- Stable URL of this page
https://metraipolis.com/documento/dc_2f38dc9b89eb
How to cite
DEPICT: Scoring Text-to-Image Alignment by Answer Agreement — arXiv:2610.03617 — arXiv (en), https://arxiv.org/abs/2610.03617 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2f38dc9b89eb (Metraipolis archive dc_2f38dc9b89eb, sha256 6b339376c1e4fa565f58dd81b4aad0777ec413eb14653fbda6cb24bd46e66583)
BibTeX
@misc{dc2f38dc9b89eb,
title = {DEPICT: Scoring Text-to-Image Alignment by Answer Agreement — arXiv:2610.03617},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.03617},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2f38dc9b89eb; sha256 6b339376c1e4fa565f58dd81b4aad0777ec413eb14653fbda6cb24bd46e66583},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.03617. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Image-text alignment is a core problem in computer vision with applications in caption evaluation, hallucination detection, data curation, and the benchmarking of text-to-image (T2I) generators. As T2I models improve, benchmarking has become demanding, requiring metrics capable of finding a series of issues like missing objects, swapped attributes, miscounts, and ignored negations. Recent work addresses this by fine-tuning evaluators on preference data or by prompting a vision-language model, either holistically with the caption or with decomposed verification questions. However, existing approaches fall short: fine-tuned metrics remain bound to one backbone and training distribution; holistic metrics miss fine-grained details; and decomposed metrics rely on a fixed-YES assumption that penalizes faithful images whenever that assumption fails. In contrast, we propose DEPICT, a training-free metric that replaces fixed reference answers with expected agreement between image-based and caption-only answers, weighting questions by how decisively the caption determines them. By replacing fixed references, our agreement rule increases negation accuracy from 19% to 88%. To recover the context lost during decomposition, DEPICT merges this agreement score with a holistic score. We evaluate DEPICT on five benchmarks and eleven backbones from three model families and find that it surpasses all training-free metrics and exceeds fine-tuned evaluators on two out of three human-correlation benchmarks. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-05 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.03617 — 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 2026-10-06T08:21:44.267Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).