# Discrete Diffusion Language Models for Interactive Radiology Report Drafting

arXiv: https://arxiv.org/abs/2607.01436 · HTML: https://mlsanity.com/paper/2607.01436/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Medical imaging and segmentation, one of 14 directions in Diffusion models.
- Builds on 21 of its 25 references that are in the map.
- First posted on arXiv 1 Jul 2026.

## Placed in

- Diffusion models / [Medical imaging and segmentation](https://mlsanity.com/diffusion/medical-imaging-and-segmentation/)

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

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Structured Denoising Diffusion Models in Discrete State-Spaces](https://mlsanity.com/paper/2107.03006/) (2021)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [Large Language Diffusion Models](https://mlsanity.com/paper/2502.09992/) (2025)
- [Simple and Effective Masked Diffusion Language Models](https://mlsanity.com/paper/2406.07524/) (2024)
- [Dream 7B: Diffusion Large Language Models](https://mlsanity.com/paper/2508.15487/) (2025)
- [SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answering](https://mlsanity.com/paper/2102.09542/) (2021)
- [MAIRA-2: Grounded Radiology Report Generation](https://mlsanity.com/paper/2406.04449/) (2024)
- [Interactive and Explainable Region-guided Radiology Report Generation](https://mlsanity.com/paper/2304.08295/) (2023)
- [MAIRA-1: A specialised large multimodal model for radiology report generation](https://mlsanity.com/paper/2311.13668/) (2023)
- [ReXrank: A Public Leaderboard for AI-Powered Radiology Report Generation](https://mlsanity.com/paper/2411.15122/) (2024)
- [DreamOn: Diffusion Language Models For Code Infilling Beyond Fixed-size Canvas](https://mlsanity.com/paper/2602.01326/) (2026)
- [ECHO: Efficient Chest X-ray Report Generation with One-step Block Diffusion](https://mlsanity.com/paper/2604.09450/) (2026)
- [SDR: Set-Distance Rewards for Radiology Report Generation](https://mlsanity.com/paper/2606.00440/) (2026)
- [AnchorDiff: Topology-Aware Masked Diffusion with Confidence-based Rewriting for Radiology Report Generation](https://mlsanity.com/paper/2605.17071/) (2026)
- [CopilotCAD: Empowering Radiologists with Report Completion Models and Quantitative Evidence from Medical Image Foundation Models](https://mlsanity.com/paper/2404.07424/) (2024)
- [Discrete Diffusion Models with MLLMs for Unified Medical Multimodal Generation](https://mlsanity.com/paper/2510.06131/) (2025)
- [SemEnrich: Self-Supervised Semantic Enrichment of Radiology Reports for Vision-Language Learning](https://mlsanity.com/paper/2604.09887/) (2026)
- [OpenMedQ: Broad Open Pretraining for Medical Vision-Language Models](https://mlsanity.com/paper/2606.12953/) (2026)

## Built on by (most cited first, 2 of 2)

- [DiffusionGemma Technical Report](https://mlsanity.com/paper/2608.00146/) (2026)
- [Intent-Triggered Anchoring for Multi-Token Format Units in Parallel Decoding of Diffusion Language Models](https://mlsanity.com/paper/2610.16516/) (2026)

## Abstract (arXiv)

Diffusion language models, which generate text by denoising a token canvas bidirectionally instead of emitting tokens left to right, have become competitive with autoregressive (AR) generation. Medical foundation models, however, remain almost entirely autoregressive. We adapt a mixture-of-experts diffusion language model, DiffusionGemma-26B, and benchmark it against its same-size AR sibling Gemma-4-26B under an identical LoRA recipe on medical visual question answering datasets, scored by a verbosity-robust LLM judge. Diffusion matches or exceeds AR on all of them, and the finetuned model (3.8B active) is competitive with frontier vision-language models; its decoding is also 3.5-4.4x faster. Beyond this parity, the diffusion model offers a drafting capability AR lacks: any-order infill. Because the canvas is denoised bidirectionally, a radiologist can fix report fragments and have the model fill the text between them, an operation inherent to diffusion but not to autoregression, which is subpar at it. This suits real reports, which are often terse or inconsistent across clinicians and institutions.