Libra: Leveraging Temporal Images for Biomedical Radiology Analysis

ACL 2025

1Information Retrieval Group
2AI4BioMed Lab
School of Computing Science, University of Glasgow, UK

πŸ”₯[NEWS!]
[4 Oct 2025] πŸ› οΈ Updated the batch inference function libra_eval_batch β€” see Batch Inference Guide.
[27 Sep 2025] πŸ“· CCD: Clinical Contrastive Decoding is out! A plug-and-play, training-free decoding framework to boost any SOTA radiology MLLM!
[20 Jul 2025] βœ… Support for LLaVA-Rad is added. Compatible weights are now available.
[15 Jul 2025] βœ… Support for MAIRA-2 is added. Compatible weights are provided for benchmarking, with use_maira_feature_norm: true set to ensure compatibility with the default feature extraction strategy.
[14 Jul 2025] 🩺 For evaluating AI-generated radiology reports, we recommend using πŸ‘‰ RadEval.
[09 Jul 2025] 🚚 The test dataset is now available on Hugging Face β€” check out ./MIMIC-CXR-RRG. It includes findings, impression, indication, comparison, technique, history, and examination sections, processed following the official MIMIC-CXR guidelines.
[08 Jul 2025] πŸ’» Data preparation scripts for Prior Image Retrieve are now available.
[16 May 2025] πŸ“ A short blog: some musings on "What Does β€˜Temporal’ Really Mean?” β€” thoughts behind Libra and temporal reasoning in radiology!
[15 May 2025] πŸ₯³ The paper has been accepted to ACL 2025!
[09 May 2025] ✨ Now with full support for the Phi-4 family β€” compact language and reasoning models from Microsoft.
[24 Mar 2025] πŸ† Libra was invited to the ReXrank Challenge β€” a leading leaderboard for Chest X-ray Report Generation.
- More Updates -
[10 Mar 2025] βœ… The architecture of LLaVA-Med v1.5 is now supported by this repo. Compatible weights are provided, with unfreeze_mm_vision_tower: true set to ensure the adapted vision encoder is used.
[11 Feb 2025] 🚨 Libra-v1.0-3b has been released! A Small Multimodal Language Model for Radiology Report Generation, following the same training strategy as Libra.
[10 Feb 2025] πŸš€ The Libra repo now supports Mistral, Phi-3, and Gemma as LLMs, along with SigLip as the encoder!
[19 Jan 2025] ⚑ The online demo is available at Hugging Face Demo. Welcome to try it out!
[07 Jan 2025] πŸ—‚οΈ The processed data is available at Data Download.
[20 Dec 2024] 🚨 Libra-v1.0-7b has been released!

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.

Demo

Explore the capabilities of Libra with our interactive demo.

Why Libra?

Temporal Awareness in Report Generation

Temporal hallucination is a critical challenge in radiology report generation (RRG). Traditional multimodal large language models (MLLMs) struggle to integrate prior images correctly, often generating:

  • Spurious references to nonexistent prior studies (Single-image case).
  • Inaccurate interpretations of disease progression (Temporal-image case).

Libra addresses these limitations by integrating a Temporal Alignment Connector (TAC) to improve temporal awareness, ensuring:

  • Prior studies are correctly referenced only when available.
  • Hallucinated references are eliminated, avoiding misleading reports.
  • Temporal changes are accurately captured, ensuring clinically meaningful outputs.

Overview

We propose Libra (Leveraging Temporal Images for Biomedical Radiology Analysis), a novel framework tailored for radiology report generation (RRG) that incorporates temporal change information to address the challenges of interpreting medical images effectively.

Libra leverages RAD-DINO, a pre-trained visual transformer, as its image encoder to generate robust and scalable image features. These features are further refined by a Temporal Alignment Connector (TAC), a key innovation in Libra's architecture. The TAC comprises:

  • Layerwise Feature Extractor (LFE): Captures high-granularity image feature embeddings from the encoder.
  • Temporal Fusion Module (TFM): Integrates temporal references from prior studies to enhance temporal awareness and reasoning.

These refined features are fed into Meditron, a specialised medical large language model (LLM), to generate comprehensive, temporally-aware radiology reports. Libra’s modular design seamlessly integrates state-of-the-art open-source pre-trained models for both image and text, aligning them through a temporal-aware adapter to ensure robust cross-modal reasoning and understanding.

Through a two-stage training strategy, Libra demonstrates the powerful potential of multimodal large language models (MLLMs) in specialised radiology applications. Extensive experiments on the MIMIC-CXR dataset highlight Libra's performance, setting a new state-of-the-art benchmark among models of the same parameter scale.

Key Contributions

  • Temporal Awareness: Libra captures and synthesizes temporal changes in medical images, addressing the challenge of handling prior study citations in RRG tasks.
  • Innovative Architecture: The Temporal Alignment Connector (TAC) ensures high-granularity feature extraction and temporal integration, significantly enhancing cross-modal reasoning capabilities.
  • State-of-the-Art Performance: Libra achieves outstanding results on the MIMIC-CXR dataset, outperforming existing MLLMs in both accuracy and temporal reasoning.
  • Libra Repository: Our code space provides a public and detailed implementation of Libra, facilitating reproducibility and further research in the field, and integrates both training and evaluation processes, ensuring a streamlined and efficient workflow for developing and testing radiology report generation models.

Experimental Results

Performance Analysis

BibTeX


@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}
}

Acknowledgement

This website is adapted from Nerfies, licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.