MedCTA Benchmark

What is MedCTA?

MedCTA evaluates clinical tool agents on 107 clinician-verified, step-implicit tasks with multimodal inputs (radiology images, pathology slides, reports) and 5…

Released

2026-06-10

Evaluates

Biology & Drug Discovery, Multimodal

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.11702

MedCTA paper, code and dataset

• Paper https://arxiv.org/abs/2606.11702
• Code https://github.com/IVUL-KAUST/MedCTA
• Dataset https://huggingface.co/datasets/IVUL-KAUST/MedCTA
• Website https://ivul-kaust.github.io/MedCTA/

Which sources cite MedCTA?

• MedCTA

Related Biology & Drug Discovery, Multimodal benchmarks

• 3DCodeBench
• A Large-Scale Dataset and Benchmark
• A-Bench
• A-OKVQA
• ABC-Bench
• ActionAtlas
• ActiveFly-Bench
• Agent Planning Benchmark