Discrete Diffusion Language Models for Interactive Radiology Report Drafting

Researchers have adapted a mixture-of-experts diffusion language model for medical applications, matching or exceeding traditional autoregressive models while decoding 3.5 to 4.4 times faster and enabling flexible, non-linear report drafting.

Computer Science > Artificial Intelligence

Title:Discrete Diffusion Language Models for Interactive Radiology Report Drafting

View PDF HTML (experimental)Abstract: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.

Source: arXiv cs.AI Recent

More in this category

agentic-systems

Janus: a Playground for User-Involved Agentic Permission Management

As AI agents autonomously execute tools, managing permissions becomes a critical challenge. Janus is introduced as a playground system consisting of Janus-Core and Janus-Harness to implement and evaluate user-involved permission management designs.

agentic-systems

Profit-Based Counterfactual Explanations for Product Improvement: A Case Study of Manga Sales in Japan

Researchers propose a novel Profit-Based Counterfactual Explanation (PBCE) framework that integrates machine learning with business profit maximization, demonstrated through a case study on Japanese manga sales.

agentic-systems

When Should Service Agents Reconsider? Difficulty-Routed Control in Customer-Service Operations

Researchers propose a difficulty-routed service-control architecture for autonomous customer-service agents. By separating routine queries from complex backend operations, the system maintains speed while preventing costly operational errors.

agentic-systems

Hawk: Harnessing Hardware-Aware Knowledge for High-Performance NPU Kernel Generation

Developing high-performance NPU kernels is a major bottleneck due to strict hardware constraints. Hawk, a training-free framework, addresses this by leveraging hardware-aware knowledge, boosting generation accuracy to 80% and achieving up to a 2.2x speedup.

agentic-systems

SemHash-LLM: A Multi-Granularity Semantic Hashing Framework for Document Deduplication

Researchers have introduced SemHash-LLM, a multi-granularity semantic hashing framework designed for efficient large-scale document deduplication. By combining LLMs with advanced hashing techniques, it reduces neural verification costs to under 1% while maintaining high accuracy.

agentic-systems

Beyond Next-Token Prediction: An RLVR Proof of Concept for Tool-Use Agents on Atlassian Workflows

Researchers demonstrate how Reinforcement Learning with Verifiable Rewards (RLVR) can bridge the gap between next-token prediction and complex API execution, significantly boosting the performance of small language models on Atlassian workflows.

EXPLORE TOPICS