Research map / Vision-language and multimodal / Human and embodied modeling
Exploring and Distilling Posterior and Prior Knowledge for Radiology Report Generation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use PPKED instead of R2Gen | Radiology report generation · MIMIC-CXR | BLEU-1: 0.353 -> 0.360 | Helped |
| Use PPKED instead of R2Gen | Radiology report generation · IU-Xray | CIDEr: PPKED 0.351; R2Gen - | Helped |
| Add PoKE to the Base HRNN (ablation) | Base HRNN · IU-Xray | CIDEr: 0.261 -> 0.285 | Helped |
| Add prior medical knowledge (PrMK) (ablation) | Base HRNN · IU-Xray | CIDEr: 0.261 -> 0.288 | Helped |
| Add prior working experience (PrWE) (ablation) | Base HRNN · IU-Xray | CIDEr: 0.261 -> 0.296 | Helped |
| Add both prior medical knowledge and prior working experience (ablation) | Base HRNN · IU-Xray | CIDEr: 0.261 -> 0.304 | Helped |
| Combine PoKE, PrMK, and PrWE (ablation) | Base HRNN · IU-Xray | CIDEr: 0.261 -> 0.311 | Helped |
| Replace LSTM decoder with Transformer Decoder (ablation) | PoKE + PrMK + PrWE · IU-Xray | CIDEr: 0.311 -> 0.337 | Helped |
| Add Adaptive Distilling Attention (ADA) (ablation) | Transformer Decoder with PoKE, PrMK, and PrWE · IU-Xray | CIDEr: 0.337 -> 0.351 | Helped |
| Reverse PoKE order from image-to-topics-to-image to topics-to-image-to-topics (ablation) | PoKE · IU-Xray | report-generation performance | Hurt |
Closest papers by meaning
- Instance-level Expert Knowledge and Aggregate Discriminative Attention for Radiology Report Generation (CVPR 2024 · similarity 0.90)
- LLM-RG4: Flexible and Factual Radiology Report Generation Across Diverse Input Contexts (AAAI 2025 · similarity 0.87)
- Bootstrapping Large Language Models for Radiology Report Generation (AAAI 2024 · similarity 0.87)
- ReFINE: A Reward-Based Framework for Interpretable and Nuanced Evaluation of Radiology Report Generation (AAAI 2026 · similarity 0.86)
- DART: Disease-aware Image-Text Alignment and Self-correcting Re-alignment for Trustworthy Radiology Report Generation (CVPR 2025 · similarity 0.86)
- Enhanced Contrastive Learning with Multi-view Longitudinal Data for Chest X-ray Report Generation (CVPR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.