CT-RAT

CT-RAT is a structured reasoning benchmark for 3D chest CT that contains diagnostic traces reflecting a radiologist's workflow, used to evaluate the reasoning capabilities of multimodal large language models in medical imaging.

Papers using CT-RAT (5)

Disease-Centric Vision-Language Pretraining with Hybrid Visual Encoding for 3D Computed Tomography2026U-VLM: Hierarchical Vision Language Modeling for Report Generation2026Curriculum-Driven 3D CT Report Generation via Language-Free Visual Grafting and Zone-Constrained Compression2026Med3D-R1: Incentivizing Clinical Reasoning in 3D Medical Vision-Language Models for Abnormality Diagnosis2026More performant and scalable: Rethinking contrastive vision-language pre-training of radiology in the LLM era2025