mlsanityData to 7 Oct 2026

Visual concept ranking uncovers medical shortcuts used by large multimodal models

arXiv 2602.05096 ↗First posted 4 Feb 2026cs.CV cs.LGcited by 0

On the map: Multimodal LLMs / vision-language models → Medical vision-language models

Abstract (arXiv)

Ensuring the reliability of machine learning models in safety-critical domains such as healthcare requires auditing methods that can uncover model shortcomings. We introduce a method for identifying important visual concepts within large multimodal models (LMMs) and use it to investigate the behaviors these models exhibit when prompted with medical tasks. We primarily focus on the task of classifying malignant skin lesions from clinical dermatology images, with supplemental experiments including both chest radiographs and natural images. After showing how LMMs display unexpected gaps in performance between different demographic subgroups when prompted with demonstrating examples, we apply our method, Visual Concept Ranking (VCR), to these models and prompts. VCR generates hypotheses related to different visual feature dependencies, which we are then able to validate with manual interventions.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 21 of its 48 references that are in the map. First posted on arXiv 4 Feb 2026.

Builds on

21 of 48 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025BiasICL: In-Context Learning and Demographic Biases of Vision Language Models4 Mar

Other references · 20

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2017Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)cited by 1,309
  3. 2023Representation Engineering: A Top-Down Approach to AI Transparencycited by 1,239
  4. 2021Probing Classifiers: Promises, Shortcomings, and Advancescited by 831
  5. 2023The Linear Representation Hypothesis and the Geometry of Large Language Modelscited by 679
  6. 2020Contrastive Learning of Medical Visual Representations from Paired Images and Textcited by 636
  7. 2023OpenFlamingo: An Open-Source Framework for Training Large Autoregressive Vision-Language Modelscited by 560
  8. 2017CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learningcited by 459
  9. 2018Confounding variables can degrade generalization performance of radiological deep learning modelscited by 374
  10. 2022CLIP-Dissect: Automatic Description of Neuron Representations in Deep Vision Networkscited by 144
  11. 2024LAB-Bench: Measuring Capabilities of Language Models for Biology Researchcited by 138
  12. 2021Disparities in Dermatology AI: Assessments Using Diverse Clinical Imagescited by 132
8 more
  1. 2024Automatically Interpreting Millions of Features in Large Language Modelscited by 106
  2. 2019Explanation by Progressive Exaggerationcited by 76
  3. 2023SkinCon: A skin disease dataset densely annotated by domain experts for fine-grained model debugging and analysiscited by 44
  4. 2023Towards Trustable Skin Cancer Diagnosis via Rewriting Model's Decisioncited by 27
  5. 2024Large Multi-modal Models Can Interpret Features in Large Multi-modal Modelscited by 23
  6. 2024The Evolution of Multimodal Model Architecturescited by 21
  7. 2024LG-CAV: Train Any Concept Activation Vector with Language Guidancecited by 9
  8. 2025Line of Sight: On Linear Representations in VLLMscited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024A Two-Step Concept-Based Approach for Enhanced Interpretability and Trust in Skin Lesion Diagnosis
  2. 2023Towards Concept-based Interpretability of Skin Lesion Diagnosis using Vision-Language Models
  3. 2026Bridging Research and Practice: A Systematic Evaluation of Generalist and Dermatology-Specific Models in Clinical Skin Lesion Classification
  4. 2025Grounding Multimodal Large Language Models with Quantitative Skin Attributes: A Retrieval Study
  5. 2026Vision-Language Models Encode Clinical Guidelines for Concept-Based Medical Reasoning
  6. 2026How Medical VLMs Underutilize Their Vision Encoders: A Dermatology Perspective
  7. 2025Citrus-V: Advancing Medical Foundation Models with Unified Medical Image Grounding for Clinical Reasoning
  8. 2026How Do Medical MLLMs Fail? A Study on Visual Grounding in Medical Images