analiza aiqbench
This arXiv preprint proposes RadCarve, a vision-language model that generates radiology report sections tied to specific anatomical regions in chest X-rays. Its main claim is that building generation units from existing region annotations, rather than from imposed structures, improves grounding between text and image. The reported gains are measured on benchmarks such as CXRFEScore, Chest ImaGenome, and PadChest-GR, and the results are not yet validated clinically. For readers, the notable points are the efficiency claim (under a sixth of the parameters of large baselines) and the grounding approach, which may transfer to other report-generation tasks.
Carving Radiology Reports at Their Evidential Joints: Anatomical Region Sets as Generation Units
arXiv:2610.29165v1 Announce Type: new Abstract: Organizing radiology reports into independently generated units offers a way to generate fine-grained clinical content and bind it to visual evidence. However, existing methods typically adopt generation-unit structures not directly specified in natural reports or their annotations, introducing additional modeling compromises. We propose RadCarve, which instead defines generation units directly from the anatomical region sets in existing statement-region correspondences, unifying unit formation with visual evidence selection. RadCarve's planner uses composition-specialized query banks to jointly predict unit existence and region composition from the image, iteratively refining both with regional evidence aggregated according to current membership predictions. The generator independently produces each unit's text conditioned on the joint visual evidence of its predicted region set. Compared with strong baselines, RadCarve improves CXRFEScore by 9.4% in unit generation given region sets and achieves 2.4 times the end-to-end binding F1. It also achieves leading report-generation results on Chest ImaGenome and PadChest-GR, using fewer than one-sixth as many parameters as large-scale baselines.
Abstract:Organizing radiology reports into independently generated units offers a way to generate fine-grained clinical content and bind it to visual evidence. However, existing methods typically adopt generation-unit structures not directly specified in natural reports or their annotations, introducing additional modeling compromises. We propose RadCarve, which instead defines generation units directly from the anatomical region sets in existing statement-region correspondences, unifying unit formation with visual evidence selection. RadCarve's planner uses composition-specialized query banks to jointly predict unit existence and region composition from the image, iteratively refining both with regional evidence aggregated according to current membership predictions. The generator independently produces each unit's text conditioned on the joint visual evidence of its predicted region set. Compared with strong baselines, RadCarve improves CXRFEScore by 9.4% in unit generation given region sets and achieves 2.4 times the end-to-end binding F1. It also achieves leading report-generation results on Chest ImaGenome and PadChest-GR, using fewer than one-sixth as many parameters as large-scale baselines.
Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2610.29165 [cs.CL]
(or arXiv:2610.29165v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2610.29165 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 258 cuvinte