Go to MIDL 2026 Conference homepageCWCD: Category-Wise Contrastive Decoding for Structured Medical Report Generation
Keywords: Radiology Report Generation, Multimodal Large Language Models, Contrastive Decoding, Chest X-rays
TL;DR: Category-Wise Contrastive Decoding (CWCD), a novel and modular framework designed to enhance structured radiology report generation.
Abstract: Interpreting chest X-rays is inherently challenging due to the overlap between anatomical structures and the subtle presentation of many clinically significant pathologies, making accurate diagnosis time-consuming even for experienced radiologists. Recent radiology-focused foundation models, such as LLaVA-Rad and Maira-2, have positioned multi-modal large language models (MLLMs) at the forefront of automated radiology report generation (RRG). However, despite these advances, current foundation models generate reports in a single forward pass. This decoding strategy diminishes attention to visual tokens and increases reliance on language priors as generation proceeds, which in turn introduces spurious pathology co-occurrences in the generated reports. To mitigate these limitations, we propose Category-Wise Contrastive Decoding (CWCD), a novel and modular framework designed to enhance structured radiology report generation (SRRG). Our approach introduces category-specific parameterization and generates category-wise reports by contrasting normal X-rays with masked X-rays using category-specific visual prompts. Experimental results demonstrate that CWCD consistently outperforms baseline methods across both clinical efficacy and natural language generation metrics. An ablation study further elucidates the contribution of each architectural component to overall performance.
Primary Subject Area: Detection and Diagnosis
Secondary Subject Area: Application: Radiology
Registration Requirement: Yes
Read CFP & Author Instructions: Yes
Originality Policy: Yes
Single-blind & Not Under Review Elsewhere: Yes
LLM Policy: Yes
Submission Number: 73
Loading