Imagen de Portada

Visual concept ranking uncovers medical shortcuts used by large multimodal models

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Janizek, Joseph D., Xu, Sonnet, Lateef, Junayd, Daneshjou, Roxana
Formato:	Preprint
Publicado:	2026
Materias:	Computer Vision and Pattern Recognition Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

BiasICL: In-Context Learning and Demographic Biases of Vision Language Models
por: Xu, Sonnet, et al.
Publicado: (2025)

Intrinsic Dimension Correlation: uncovering nonlinear connections in multimodal representations
por: Basile, Lorenzo, et al.
Publicado: (2024)

Closing the gap in multimodal medical representation alignment
por: Grassucci, Eleonora, et al.
Publicado: (2026)

Confidence intervals uncovered: Are we ready for real-world medical imaging AI?
por: Christodoulou, Evangelia, et al.
Publicado: (2024)

Human-like object concept representations emerge naturally in multimodal large language models
por: Du, Changde, et al.
Publicado: (2024)

Bayesian computation with generative diffusion models by Multilevel Monte Carlo
por: Haji-Ali, Abdul-Lateef, et al.
Publicado: (2024)

Explaining latent representations of generative models with large multimodal models
por: Zhu, Mengdan, et al.
Publicado: (2024)

Bridging visual saliency and large language models for explainable deep learning in medical imaging
por: Nguezet, Paul Valery, et al.
Publicado: (2026)

A multimodal slice discovery framework for systematic failure detection and explanation in medical image classification
por: Liu, Yixuan, et al.
Publicado: (2026)

UCSF-PDGM-VQA: Visual Question Answering dataset for brain tumor MRI interpretation
por: Ghosh, Shiv, et al.
Publicado: (2026)

Concept-TRAK: Understanding how diffusion models learn concepts through concept-level attribution
por: Park, Yonghyun, et al.
Publicado: (2025)

Visual hallucination detection in large vision-language models via evidential conflict
por: Huang, Tao, et al.
Publicado: (2025)

How can embedding models bind concepts?
por: Uselis, Arnas, et al.
Publicado: (2026)

Performance uncertainty in medical image analysis: a large-scale investigation of confidence intervals
por: André, Pascaline, et al.
Publicado: (2026)

MULTIAQUA: A multimodal maritime dataset and robust training strategies for multimodal semantic segmentation
por: Muhovič, Jon, et al.
Publicado: (2025)

What do vision-language models see in the context? Investigating multimodal in-context learning
por: Santos, Gabriel O. dos, et al.
Publicado: (2025)

ImageNot: A contrast with ImageNet preserves model rankings
por: Salaudeen, Olawale, et al.
Publicado: (2024)

Visual representations in the human brain are aligned with large language models
por: Doerig, Adrien, et al.
Publicado: (2022)

Skull-stripping induces shortcut learning in MRI-based Alzheimer's disease classification
por: Tinauer, Christian, et al.
Publicado: (2025)

Do multimodal models imagine electric sheep?
por: Ramakrishnan, Santhosh Kumar, et al.
Publicado: (2026)

AG-Fusion: adaptive gated multimodal fusion for 3d object detection in complex scenes
por: Liu, Sixian, et al.
Publicado: (2025)

How to build the best medical image segmentation algorithm using foundation models: a comprehensive empirical study with Segment Anything Model
por: Gu, Hanxue, et al.
Publicado: (2024)

Bridging The Gap between Low-rank and Orthogonal Adaptation via Householder Reflection Adaptation
por: Yuan, Shen, et al.
Publicado: (2024)

Soft-CAM: Making black box models self-explainable for medical image analysis
por: Djoumessi, Kerol, et al.
Publicado: (2025)

Classification of freshwater snails of the genus Radomaniola with multimodal triplet networks
por: Vetter, Dennis, et al.
Publicado: (2024)

A comprehensive and easy-to-use multi-domain multi-task medical imaging meta-dataset
por: Woerner, Stefano, et al.
Publicado: (2024)

Selective experience replay compression using coresets for lifelong deep reinforcement learning in medical imaging
por: Zheng, Guangyao, et al.
Publicado: (2023)

Buffer replay enhances the robustness of multimodal learning under missing-modality
por: Zhu, Hongye, et al.
Publicado: (2025)

Can multimodal representation learning by alignment preserve modality-specific information?
por: Thoreau, Romain, et al.
Publicado: (2025)

Do ImageNet-trained models learn shortcuts? The impact of frequency shortcuts on generalization
por: Wang, Shunxin, et al.
Publicado: (2025)

Towards a multimodal framework for remote sensing image change retrieval and captioning
por: Ferrod, Roger, et al.
Publicado: (2024)

ReasonEdit: Editing Vision-Language Models using Human Reasoning
por: Qiu, Jiaxing, et al.
Publicado: (2026)

On the effectiveness of multimodal privileged knowledge distillation in two vision transformer based diagnostic applications
por: Baur, Simon, et al.
Publicado: (2025)

Sparse autoencoders reveal selective remapping of visual concepts during adaptation
por: Lim, Hyesu, et al.
Publicado: (2024)

COOD: Combined out-of-distribution detection using multiple measures for anomaly & novel class detection in large-scale hierarchical classification
por: Hogeweg, L. E., et al.
Publicado: (2024)

DGSSM: Diffusion guided state-space models for multimodal salient object detection
por: Ghosh, Suklav, et al.
Publicado: (2026)

How Visual Representations Map to Language Feature Space in Multimodal LLMs
por: Venhoff, Constantin, et al.
Publicado: (2025)

Aligning Visual Contrastive learning models via Preference Optimization
por: Afzali, Amirabbas, et al.
Publicado: (2024)

Visual Pre-Training on Unlabeled Images using Reinforcement Learning
por: Ghosh, Dibya, et al.
Publicado: (2025)

Self-supervised cost of transport estimation for multimodal path planning
por: Gherold, Vincent, et al.
Publicado: (2024)