A Small Multimodal Language Model for Radiology Report Generation, following the same training strategy as Libra. It can produce detailed Findings sections with temporal comparisons (e.g., comparing a current chest X-ray with prior ones). Libra integrates the following key components:

RAD-DINO: A vision encoder pre-trained on medical imaging datasets for robust feature extraction from chest X-rays.
Llama-3.2-3B-Instruct: A 3B-parameter language model (based on Llama-3.2). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks.
Temporal Alignment Connector (TAC): A custom adapter that fuses features across multiple time points to enable temporal comparisons.

This model card provides an overview of Libra’s architecture, training methodology, limitations, and recommended usage guidelines.

Paper and Resources

For more detailed information regarding Libra’s methodology, theoretical foundation, and performance benchmarks, please refer to the following resources:

Project Website: Libra v1.0
Paper: arXiv:2411.19378
Code Repository: X-iZhang/Libra (GitHub)

Or check out our Spaces demo!

Training Strategy

Libra is trained in a two-stage process:

Temporal Feature Alignment
- Trains TAC to effectively fuse and align features from different time points (current and previous chest X-rays).
- Focuses on capturing notable changes (e.g., appearance or progression of opacities, devices, and lines).
Fine-Tuning for Radiology Report Generation
- The language model part is fine-tuned on a large dataset of paired chest X-ray images and radiology reports.
- Emphasizes the generation of the Findings section, especially incorporating temporal descriptors.

Intended Use

Libra is primarily designed to assist clinical practitioners, researchers, and medical students in generating chest X-ray reports. Key applications include:

Clinical Decision Support: Providing draft findings that can be refined by a radiologist.
Educational Tool: Demonstrating example interpretations and temporal changes for training radiology residents.
Research: Facilitating studies on automated report generation and temporal feature learning in medical imaging.

Important: Outputs should be reviewed by qualified radiologists or medical professionals before final clinical decisions are made.

Limitations and Recommendations

Data Bias: The model’s performance may be less reliable for underrepresented demographics or rare pathologies.
Clinical Oversight: Always involve a medical professional to verify the results—Libra is not a substitute for professional judgment.
Temporal Inaccuracies: Despite TAC’s focus on temporal alignment, subtle or uncommon changes may go unrecognized.
Generalization: Libra’s performance on chest X-ray types or conditions not seen during training may be limited.

Ethical Considerations

Patient Privacy: Ensure the data is fully de-identified and compliant with HIPAA/GDPR (or relevant privacy regulations).
Responsible Use: Deploy Libra’s outputs carefully; they are not guaranteed to be error-free.
Accountability: Users and organizations must assume responsibility for verifying clinical accuracy and safety.

How to Cite ✒️

If you use Libra in academic or research contexts, please cite:

bibtex

@inproceedings{zhang-etal-2025-libra,
    title = "Libra: Leveraging Temporal Images for Biomedical Radiology Analysis",
    author = "Zhang, Xi  and
      Meng, Zaiqiao  and
      Lever, Jake  and
      Ho, Edmond S. L.",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.findings-acl.888/",
    pages = "17275--17303",
    ISBN = "979-8-89176-256-5",
    abstract = "Radiology report generation (RRG) requires advanced medical image analysis, effective temporal reasoning, and accurate text generation. While multimodal large language models (MLLMs) align with pre-trained vision encoders to enhance visual-language understanding, most existing methods rely on single-image analysis or rule-based heuristics to process multiple images, failing to fully leverage temporal information in multi-modal medical datasets. In this paper, we introduce **Libra**, a temporal-aware MLLM tailored for chest X-ray report generation. Libra combines a radiology-specific image encoder with a novel Temporal Alignment Connector (**TAC**), designed to accurately capture and integrate temporal differences between paired current and prior images. Extensive experiments on the MIMIC-CXR dataset demonstrate that Libra establishes a new state-of-the-art benchmark among similarly scaled MLLMs, setting new standards in both clinical relevance and lexical accuracy. All source code and data are publicly available at: https://github.com/X-iZhang/Libra."
}

@inproceedings{zhang2025libra,
  title={Libra: Leveraging temporal images for biomedical radiology analysis},
  author={Zhang, Xi and Meng, Zaiqiao and Lever, Jake and Ho, Edmond SL},
  booktitle={Findings of the Association for Computational Linguistics: ACL 2025},
  pages={17275--17303},
  year={2025}
}

Disclaimer:

This tool is for research and educational purposes only. It is not FDA-approved or CE-marked for clinical use. Users should consult qualified healthcare professionals for any clinical decisions.

X-iZhang/libra-v1.0-3b

Libra Model Card

Overview