Multimodal large language models (MLLMs) are increasingly deployed in real-world, agentic settings where outputs must not only be correct, but also conform to predefined data schemas. Despite recent progress in structured generation in textual domain, there is still no benchmark that systematically evaluates schema-grounded information extraction and reasoning over visual inputs. In this work, we conduct a comprehensive study of visual structural output capabilities for MLLMs with our carefully designed SO-Bench benchmark. Covering four visual domains, including UI screens, natural images, documents, and charts, SO-Bench is built from over 6.5K diverse JSON schemas and 1.8K curated image-schema pairs with human-verified quality. Benchmarking experiments on open-sourced and frontier proprietary models reveal persistent gaps in predicting accurate, schema compliant outputs, highlighting the need for better multimodal structured reasoning. Beyond benchmarking, we further conduct training experiments to largely improve the model's structured output capability. We make the benchmark and evaluation publicly available at https://github.com/apple/ml-sobench
@article{arxiv.2511.21750,
title = {SO-Bench: A Structural Output Evaluation of Multimodal LLMs},
author = {Di Feng and Kaixin Ma and Feng Nan and Haofeng Chen and Bohan Zhai and David Griffiths and Mingfei Gao and Zhe Gan and Eshan Verma and Yinfei Yang and Zhifeng Chen and Afshin Dehghan},
journal= {arXiv preprint arXiv:2511.21750},
year = {2026}
}
Comments
v3 preprint. Added the link to the public benchmark