CWCD: Category-Wise Contrastive Decoding for Structured Medical Report Generation
For radiologists and automated diagnosis, CWCD reduces false pathology associations in generated reports, improving reliability of structured report generation.
The paper tackles the problem of spurious pathology co-occurrences in radiology report generation by multi-modal LLMs. The proposed Category-Wise Contrastive Decoding (CWCD) framework improves clinical efficacy and NLG metrics over baselines, with ablation studies confirming component contributions.
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.