Evaluating Cell Type Inference in Vision Language Models Under Varying Visual Context

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Singhal, Samarth, Singhal, Sandeep
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913893994987520
author Singhal, Samarth
Singhal, Sandeep
author_facet Singhal, Samarth
Singhal, Sandeep
contents Vision-Language Models (VLMs) have rapidly advanced alongside Large Language Models (LLMs). This study evaluates the capabilities of prominent generative VLMs, such as GPT-4.1 and Gemini 2.5 Pro, accessed via APIs, for histopathology image classification tasks, including cell typing. Using diverse datasets from public and private sources, we apply zero-shot and one-shot prompting methods to assess VLM performance, comparing them against custom-trained Convolutional Neural Networks (CNNs). Our findings demonstrate that while one-shot prompting significantly improves VLM performance over zero-shot ($p \approx 1.005 \times 10^{-5}$ based on Kappa scores), these general-purpose VLMs currently underperform supervised CNNs on most tasks. This work underscores both the promise and limitations of applying current VLMs to specialized domains like pathology via in-context learning. All code and instructions for reproducing the study can be accessed from the repository https://www.github.com/a12dongithub/VLMCCE.
format Preprint
id arxiv_https___arxiv_org_abs_2506_12683
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Evaluating Cell Type Inference in Vision Language Models Under Varying Visual Context
Singhal, Samarth
Singhal, Sandeep
Computer Vision and Pattern Recognition
Quantitative Methods
Vision-Language Models (VLMs) have rapidly advanced alongside Large Language Models (LLMs). This study evaluates the capabilities of prominent generative VLMs, such as GPT-4.1 and Gemini 2.5 Pro, accessed via APIs, for histopathology image classification tasks, including cell typing. Using diverse datasets from public and private sources, we apply zero-shot and one-shot prompting methods to assess VLM performance, comparing them against custom-trained Convolutional Neural Networks (CNNs). Our findings demonstrate that while one-shot prompting significantly improves VLM performance over zero-shot ($p \approx 1.005 \times 10^{-5}$ based on Kappa scores), these general-purpose VLMs currently underperform supervised CNNs on most tasks. This work underscores both the promise and limitations of applying current VLMs to specialized domains like pathology via in-context learning. All code and instructions for reproducing the study can be accessed from the repository https://www.github.com/a12dongithub/VLMCCE.
title Evaluating Cell Type Inference in Vision Language Models Under Varying Visual Context
topic Computer Vision and Pattern Recognition
Quantitative Methods
url https://arxiv.org/abs/2506.12683