# Random Direct Preference Optimization for Radiography Report Generation

arXiv: https://arxiv.org/abs/2509.21351 · HTML: https://mlsanity.com/paper/2509.21351/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Medical LLM alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 21 of its 30 references that are in the map.
- First posted on arXiv 19 Sep 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Medical LLM alignment](https://mlsanity.com/alignment/medical-llm-alignment/)

## Builds on (most cited first, 21 of 21)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Swin Transformer V2: Scaling Up Capacity and Resolution](https://mlsanity.com/paper/2111.09883/) (2021)
- [CvT: Introducing Convolutions to Vision Transformers](https://mlsanity.com/paper/2103.15808/) (2021)
- [RadGraph: Extracting Clinical Entities and Relations from Radiology Reports](https://mlsanity.com/paper/2106.14463/) (2021)
- [PadChest: A large chest x-ray image dataset with multi-label annotated reports](https://mlsanity.com/paper/1901.07441/) (2019)
- [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://mlsanity.com/paper/1907.12461/) (2019)
- [MAIRA-2: Grounded Radiology Report Generation](https://mlsanity.com/paper/2406.04449/) (2024)
- [Clinically Accurate Chest X-Ray Report Generation](https://mlsanity.com/paper/1904.02633/) (2019)
- [CheXpert Plus: Augmenting a Large Chest X-ray Dataset with Text Radiology Reports, Patient Demographics and Additional Image Formats](https://mlsanity.com/paper/2405.19538/) (2024)
- [Improving Chest X-Ray Report Generation by Leveraging Warm Starting](https://mlsanity.com/paper/2201.09405/) (2022)
- [BIMCV COVID-19+: a large annotated dataset of RX and CT images from COVID-19 patients](https://mlsanity.com/paper/2006.01174/) (2020)
- [ReXrank: A Public Leaderboard for AI-Powered Radiology Report Generation](https://mlsanity.com/paper/2411.15122/) (2024)
- [Medical Image Captioning via Generative Pretrained Transformers](https://mlsanity.com/paper/2209.13983/) (2022)
- [Overview of the First Shared Task on Clinical Text Generation: RRG24 and "Discharge Me!"](https://mlsanity.com/paper/2409.16603/) (2024)
- [Producing radiologist-quality reports for interpretable artificial intelligence](https://mlsanity.com/paper/1806.00340/) (2018)
- [e-Health CSIRO at RRG24: Entropy-Augmented Self-Critical Sequence Training for Radiology Report Generation](https://mlsanity.com/paper/2408.03500/) (2024)

## Built on by (most cited first, 1 of 1)

- [Maximizing Mutual Information Between Prompt and Response Improves LLM Performance With No Additional Data](https://mlsanity.com/paper/2603.19294/) (2026)

## Abstract (arXiv)

Radiography Report Generation (RRG) has gained significant attention in medical image analysis as a promising tool for alleviating the growing workload of radiologists. However, despite numerous advancements, existing methods have yet to achieve the quality required for deployment in real-world clinical settings. Meanwhile, large Visual Language Models (VLMs) have demonstrated remarkable progress in the general domain by adopting training strategies originally designed for Large Language Models (LLMs), such as alignment techniques. In this paper, we introduce a model-agnostic framework to enhance RRG accuracy using Direct Preference Optimization (DPO). Our approach leverages random contrastive sampling to construct training pairs, eliminating the need for reward models or human preference annotations. Experiments on supplementing three state-of-the-art models with our Random DPO show that our method improves clinical performance metrics by up to 5%, without requiring any additional training data.