# Forensic Reproducibility Audit of a Radiology Vision-Language Model Benchmark: From Intended Protocol to Released Artifact

arXiv: https://arxiv.org/abs/2607.25589 · HTML: https://mlsanity.com/paper/2607.25589/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 6 of its 27 references that are in the map.
- First posted on arXiv 28 Jul 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)

## Builds on (most cited first, 6 of 6)

- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [Datasheets for Datasets](https://mlsanity.com/paper/1803.09010/) (2018)
- [Model Cards for Model Reporting](https://mlsanity.com/paper/1810.03993/) (2018)
- [Improving Reproducibility in Machine Learning Research (A Report from the NeurIPS 2019 Reproducibility Program)](https://mlsanity.com/paper/2003.12206/) (2020)
- [Best Practices for Large Language Models in Radiology](https://mlsanity.com/paper/2412.01233/) (2024)
- [Ten Simple Rules for Making Research Software More Robust](https://mlsanity.com/paper/1610.04546/) (2016)

## Abstract (arXiv)

Medical-imaging AI benchmarks combine datasets, DICOM rendering, prompts, provider APIs, automated labels, statistical code, manuscripts, and repository releases. Agreement across these artifacts is usually assumed rather than tested. We performed a retrospective forensic reproducibility audit of a preserved chest-radiograph vision-language model (VLM) pilot; no model was called again and no image or report was newly annotated. We traced prompt bindings, DICOM metadata, output completeness, label extraction, matched analyses, and release propagation. Of 300 planned model-prompt calls, 297 yielded nonempty reports. Sixty Claude calls labeled A/B were executed with the same C prompt. The 30 studies represented 28 patients. Four MONOCHROME1 images were rendered without required polarity inversion, dataset split membership was not retained, and the unvalidated extractor truncated five reports to 4000 characters. Reconstructing one common cohort of 369 complete case-finding blocks changed Cochran's Q from 154.73 to 182.29. Of 45 McNemar comparisons, 27 had unadjusted p < 0.05 and 20 remained below 0.05 after Holm adjustment. These values describe only the archived automated-label matrix; they do not recover the intended prompt comparison or establish clinical performance. We withdraw the original performance, ranking, prompt-effect, and clinical claims and specify machine-verifiable controls for cohort, DICOM rendering, prompt and model identity, call status, annotation provenance, keyed analysis, and derived artifacts.