# Unify, Align and Refine: Multi-Level Semantic Alignment for Radiology Report Generation

arXiv: https://arxiv.org/abs/2303.15932 · HTML: https://mlsanity.com/paper/2303.15932/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 52 of its 74 references that are in the map.
- First posted on arXiv 28 Mar 2023.

## Builds on (most cited first, 30 of 52)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling](https://mlsanity.com/paper/1412.3555/) (2014)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [Microsoft COCO Captions: Data Collection and Evaluation Server](https://mlsanity.com/paper/1504.00325/) (2015)
- [Show, Attend and Tell: Neural Image Caption Generation with Visual Attention](https://mlsanity.com/paper/1502.03044/) (2015)
- [Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering](https://mlsanity.com/paper/1707.07998/) (2017)
- [Deep Visual-Semantic Alignments for Generating Image Descriptions](https://mlsanity.com/paper/1412.2306/) (2014)
- [Show and Tell: A Neural Image Caption Generator](https://mlsanity.com/paper/1411.4555/) (2014)
- [GLM-130B: An Open Bilingual Pre-trained Model](https://mlsanity.com/paper/2210.02414/) (2022)
- [MIMIC-CXR-JPG, a large publicly available database of labeled chest radiographs](https://mlsanity.com/paper/1901.07042/) (2019)
- [Self-critical Sequence Training for Image Captioning](https://mlsanity.com/paper/1612.00563/) (2016)
- [Unified Vision-Language Pre-Training for Image Captioning and VQA](https://mlsanity.com/paper/1909.11059/) (2019)
- [Long-term Recurrent Convolutional Networks for Visual Recognition and Description](https://mlsanity.com/paper/1411.4389/) (2014)
- [Generating Radiology Reports via Memory-driven Transformer](https://mlsanity.com/paper/2010.16056/) (2020)
- [Meshed-Memory Transformer for Image Captioning](https://mlsanity.com/paper/1912.08226/) (2019)
- [Knowing When to Look: Adaptive Attention via A Visual Sentinel for Image Captioning](https://mlsanity.com/paper/1612.01887/) (2016)
- [Scan2Cap: Context-aware Dense Captioning in RGB-D Scans](https://mlsanity.com/paper/2012.02206/) (2020)
- [On the Automatic Generation of Medical Imaging Reports](https://mlsanity.com/paper/1711.08195/) (2017)
- [Cross-modal Memory Networks for Radiology Report Generation](https://mlsanity.com/paper/2204.13258/) (2022)
- [DenseCap: Fully Convolutional Localization Networks for Dense Captioning](https://mlsanity.com/paper/1511.07571/) (2015)
- [Exploring and Distilling Posterior and Prior Knowledge for Radiology Report Generation](https://mlsanity.com/paper/2106.06963/) (2021)

## Built on by (most cited first, 23 of 23)

- [Automated Radiology Report Generation: A Review of Recent Advances](https://mlsanity.com/paper/2405.10842/) (2024)
- [Enhancing Representation in Radiography-Reports Foundation Model: A Granular Alignment Algorithm Using Masked Contrastive Learning](https://mlsanity.com/paper/2309.05904/) (2023)
- [Complex Organ Mask Guided Radiology Report Generation](https://mlsanity.com/paper/2311.02329/) (2023)
- [G2L: Semantically Aligned and Uniform Video Grounding via Geodesic and Game Theory](https://mlsanity.com/paper/2307.14277/) (2023)
- [Has Multimodal Learning Delivered Universal Intelligence in Healthcare? A Comprehensive Survey](https://mlsanity.com/paper/2408.12880/) (2024)
- [Multimodal Large Language Model driven Radiology Report Generation with Clinical Knowledge Enhancement](https://mlsanity.com/paper/2403.06728/) (2024)
- [Customizing General-Purpose Foundation Models for Medical Report Generation](https://mlsanity.com/paper/2306.05642/) (2023)
- [MedRAT: Unpaired Medical Report Generation via Auxiliary Tasks](https://mlsanity.com/paper/2407.03919/) (2024)
- [Pre-training on High Definition X-ray Images: An Experimental Study](https://mlsanity.com/paper/2404.17926/) (2024)
- [A Survey of Deep Learning-based Radiology Report Generation Using Multimodal Data](https://mlsanity.com/paper/2405.12833/) (2024)
- [Global Contrastive Training for Multimodal Electronic Health Records with Language Supervision](https://mlsanity.com/paper/2404.06723/) (2024)
- [Image-aware Evaluation of Generated Medical Reports](https://mlsanity.com/paper/2410.17357/) (2024)
- [Generation of Radiology Findings in Chest X-Ray by Leveraging Collaborative Knowledge](https://mlsanity.com/paper/2306.10448/) (2023)
- [MedCycle: Unpaired Medical Report Generation via Cycle-Consistency](https://mlsanity.com/paper/2403.13444/) (2024)
- [Automated Structured Radiology Report Generation with Rich Clinical Context](https://mlsanity.com/paper/2510.00428/) (2025)
- [AnchorDiff: Topology-Aware Masked Diffusion with Confidence-based Rewriting for Radiology Report Generation](https://mlsanity.com/paper/2605.17071/) (2026)
- [Improving Medical Visual Representation Learning with Pathological-level Cross-Modal Alignment and Correlation Exploration](https://mlsanity.com/paper/2506.10573/) (2025)
- [RIHA: Report-Image Hierarchical Alignment for Radiology Report Generation](https://mlsanity.com/paper/2604.27559/) (2026)
- [I2KD-SLU: An Intra-Inter Knowledge Distillation Framework for Zero-Shot Cross-Lingual Spoken Language Understanding](https://mlsanity.com/paper/2310.02594/) (2023)
- [LLaMA-XR: A Novel Framework for Radiology Report Generation using LLaMA and QLoRA Fine Tuning](https://mlsanity.com/paper/2506.03178/) (2025)
- [Medical Report Generation: A Hierarchical Task Structure-Based Cross-Modal Causal Intervention Framework](https://mlsanity.com/paper/2511.02271/) (2025)
- [Spatio-Temporal and Clinical Conditioning for Fine-Grained Radiology Report Retrieval](https://mlsanity.com/paper/2607.02024/) (2026)
- [A multicenter benchmark and clinically structured metric for coronary CTA report generation](https://mlsanity.com/paper/2609.00909/) (2026)

## Abstract (arXiv)

Automatic radiology report generation has attracted enormous research interest due to its practical value in reducing the workload of radiologists. However, simultaneously establishing global correspondences between the image (e.g., Chest X-ray) and its related report and local alignments between image patches and keywords remains challenging. To this end, we propose an Unify, Align and then Refine (UAR) approach to learn multi-level cross-modal alignments and introduce three novel modules: Latent Space Unifier (LSU), Cross-modal Representation Aligner (CRA) and Text-to-Image Refiner (TIR). Specifically, LSU unifies multimodal data into discrete tokens, making it flexible to learn common knowledge among modalities with a shared network. The modality-agnostic CRA learns discriminative features via a set of orthonormal basis and a dual-gate mechanism first and then globally aligns visual and textual representations under a triplet contrastive loss. TIR boosts token-level local alignment via calibrating text-to-image attention with a learnable mask. Additionally, we design a two-stage training procedure to make UAR gradually grasp cross-modal alignments at different levels, which imitates radiologists' workflow: writing sentence by sentence first and then checking word by word. Extensive experiments and analyses on IU-Xray and MIMIC-CXR benchmark datasets demonstrate the superiority of our UAR against varied state-of-the-art methods.