# SHOVIR: A Benchmark for Evaluating Vision Shortcut Learning in Radiology Report Generation

arXiv: https://arxiv.org/abs/2606.30201 · HTML: https://mlsanity.com/paper/2606.30201/

## 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 32 of its 47 references that are in the map.
- First posted on arXiv 29 Jun 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, 30 of 32)

- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [MedGemma Technical Report](https://mlsanity.com/paper/2507.05201/) (2025)
- [Hallucination of Multimodal Large Language Models: A Survey](https://mlsanity.com/paper/2404.18930/) (2024)
- [Confounding variables can degrade generalization performance of radiological deep learning models](https://mlsanity.com/paper/1807.00431/) (2018)
- [Generating Radiology Reports via Memory-driven Transformer](https://mlsanity.com/paper/2010.16056/) (2020)
- [A Survey on Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2402.00253/) (2024)
- [CheXbert: Combining Automatic Labelers and Expert Annotations for Accurate Radiology Report Labeling Using BERT](https://mlsanity.com/paper/2004.09167/) (2020)
- [RadGraph: Extracting Clinical Entities and Relations from Radiology Reports](https://mlsanity.com/paper/2106.14463/) (2021)
- [On the Automatic Generation of Medical Imaging Reports](https://mlsanity.com/paper/1711.08195/) (2017)
- [Hidden Stratification Causes Clinically Meaningful Failures in Machine Learning for Medical Imaging](https://mlsanity.com/paper/1909.12475/) (2019)
- [MAIRA-2: Grounded Radiology Report Generation](https://mlsanity.com/paper/2406.04449/) (2024)
- [Interactive and Explainable Region-guided Radiology Report Generation](https://mlsanity.com/paper/2304.08295/) (2023)
- [GREEN: Generative Radiology Report Evaluation and Error Notation](https://mlsanity.com/paper/2405.03595/) (2024)
- [A Whac-A-Mole Dilemma: Shortcuts Come in Multiples Where Mitigating One Amplifies Others](https://mlsanity.com/paper/2212.04825/) (2022)
- [CARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language Models](https://mlsanity.com/paper/2406.06007/) (2024)
- [Detecting and Evaluating Medical Hallucinations in Large Vision Language Models](https://mlsanity.com/paper/2406.10185/) (2024)
- [Detecting Shortcut Learning for Fair Medical AI using Shortcut Testing](https://mlsanity.com/paper/2207.10384/) (2022)
- [VLind-Bench: Measuring Language Priors in Large Vision-Language Models](https://mlsanity.com/paper/2406.08702/) (2024)
- [PadChest-GR: A Bilingual Chest X-ray Dataset for Grounded Radiology Report Generation](https://mlsanity.com/paper/2411.05085/) (2024)
- [Libra: Leveraging Temporal Images for Biomedical Radiology Analysis](https://mlsanity.com/paper/2411.19378/) (2024)
- [ReXrank: A Public Leaderboard for AI-Powered Radiology Report Generation](https://mlsanity.com/paper/2411.15122/) (2024)
- [RadVLM: A Multitask Conversational Vision-Language Model for Radiology](https://mlsanity.com/paper/2502.03333/) (2025)
- [Detecting Shortcuts in Medical Images -- A Case Study in Chest X-rays](https://mlsanity.com/paper/2211.04279/) (2022)
- [Localizing Before Answering: A Hallucination Evaluation Benchmark for Grounded Medical Multimodal LLMs](https://mlsanity.com/paper/2505.00744/) (2025)
- [e-Health CSIRO at RRG24: Entropy-Augmented Self-Critical Sequence Training for Radiology Report Generation](https://mlsanity.com/paper/2408.03500/) (2024)
- [The Impact of Auxiliary Patient Data on Automated Chest X-Ray Report Generation and How to Incorporate It](https://mlsanity.com/paper/2406.13181/) (2024)
- [Reasoning Visual Language Model for Chest X-Ray Analysis](https://mlsanity.com/paper/2510.23968/) (2025)

## Abstract (arXiv)

Current evaluation protocols for Vision-Language Models (VLMs) in Radiology Report Generation (RRG) rely on report-level metrics that measure lexical overlap or aggregate clinical correctness. However, such metrics do not test whether individual diagnostic statements stem from the actual pathological evidence visible in the image. This allows models to achieve competitive scores by exploiting learned priors or spurious correlations, a failure mode we refer to as vision shortcut. We introduce SHOVIR, a benchmark for evaluating vision shortcut behavior in RRG. SHOVIR extends two spatially annotated chest X-ray datasets, MIMIC-CXR and PadChest-GR, with per-box CheXpert labels, and defines image-level and disease-level occlusion experiments that contrast baseline performance on clean images against localized, region-specific perturbations. Comparing predictions across these conditions isolates two failure modes at the disease-class level: direct shortcuts, where a finding persists after its visual evidence is removed, and contextual shortcuts, where detection degrades once co-occurring pathologies are occluded despite the target region remaining intact. Benchmarking eight state-of-the-art VLMs, we find that shortcut behavior varies substantially across architectures and datasets. Models achieving the highest baseline report quality do not necessarily rank highest in spatial grounding, revealing that clinically fluent generation can coexist with shallow reliance on visual evidence. These findings expose a blind spot in current RRG evaluation and motivate region-aware assessment protocols.