LIMITR: Leveraging Local Information for Medical Image-Text Representation

Gefen Dawidowicz, Elad Hirsch, A. Tal
2/14/2026

No official code found. Best implementation: gefend/LIMITR (17 stars, last commit 2024-02-08).

Abstract

Medical imaging analysis plays a critical role in the diagnosis and treatment of various medical conditions. This paper focuses on chest X-ray images and their corresponding radiological reports. It presents a new model that learns a joint X-ray image & report representation. The model is based on a novel alignment scheme between the visual data and the text, which takes into account both local and global information. Furthermore, the model integrates domain-specific information of two types—lateral images and the consistent visual structure of chest images. Our representation is shown to benefit three types of retrieval tasks: text-image retrieval, class-based retrieval, and phrase-grounding. Our code is publicly available1.

View on arXivView PDF

Code Implementations(1)

gefend/LIMITRCommunity57%

Implementation of the paper LIMITR: Leveraging Local Information for Medical Image-Text Representation

173PythonJul 17, 20232 years ago

Similar Papers

Bootstrapping Chest CT Image Understanding by Distilling Knowledge from X-Ray Expert Models

87% match2026

A Disease-Aware Dual-Stage Framework for Chest X-ray Report Generation

85% match2025

CARZero: Cross-Attention Alignment for Radiology Zero-Shot Classification

85% match1 repo732026

EIR: Enhanced Image Representations for Medical Report Generation

85% match2025

MedCycle: Unpaired Medical Report Generation via Cycle-Consistency

85% match2024

3D-CT-GPT: Generating 3D Radiology Reports through Integration of Large Vision-Language Models

85% match2024

Cite this paper

@article{dawidowicz2026limitr,
  title  = {LIMITR: Leveraging Local Information for Medical Image-Text Representation},
  author = {Gefen Dawidowicz and Elad Hirsch and A. Tal},
  year   = {2026},
  eprint = {2303.11755},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2303.11755},
  journal = {ICCV 2023 2023}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2303.11755.svg)](https://codeofpaper.com/paper/2303.11755)

Discussion