Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Hybrid Retrieval-Generation Reinforced Agent for Medical Image Report Generation
Read the paperCheck 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 |
|---|---|---|---|
| combine template retrieval and sentence generation under hierarchical RL (ablation) | HRGR-Agent · CX-CHR | CIDEr: 2.800 -> 2.895 | Helped |
| add RL fine-tuning to the hybrid retrieval-generation model (ablation) | HRG · CX-CHR | CIDEr: 2.800 -> 2.895; increases CIDEr score by 0.73 compared to HRG | Helped |
| add template retrieval to a generation-only model (ablation) | Generation · CX-CHR | abnormality terminology precision (%): 27.50 -> 29.19 | Helped |
| use template retrieval without sentence generation (ablation) | HRGR-Agent · CX-CHR | CIDEr: 0.361 -> 2.565 | Helped |
| use template retrieval without sentence generation (ablation) | HRGR-Agent · CX-CHR | abnormality terminology precision (%): 27.50 -> 14.13 | Hurt |
| remove the generation module and rely on retrieved templates (ablation) | HRGR-Agent · CX-CHR | CIDEr: 2.895 -> 2.565 | Hurt |
| remove both template retrieval and RL fine-tuning (ablation) | HRGR-Agent · CX-CHR | CIDEr: 2.895 -> 0.361 | Hurt |
| use HRGR-Agent rather than CoAtt | medical report generation · IU X-Ray | CIDEr: 0.277 -> 0.343 | Helped |
| use HRGR-Agent rather than CoAtt | medical report generation · IU X-Ray | BLEU-1: 0.455 -> 0.438 | Hurt |
| use HRGR-Agent rather than CoAtt | medical report generation · IU X-Ray | BLEU-2: 0.288 -> 0.298 | Helped |
| use HRGR-Agent rather than CoAtt | medical report generation · IU X-Ray | BLEU-4: 0.154 -> 0.151 | Hurt |
| use HRGR-Agent rather than CoAtt | medical report generation · IU X-Ray | ROUGE: 0.369 -> 0.322 | Hurt |
Closest papers by meaning
- AHIVE: Anatomy-aware Hierarchical Vision Encoding for Interactive Radiology Report Retrieval (CVPR 2024 · similarity 0.85)
- Patho-AgenticRAG: Towards Multimodal Agentic Retrieval-Augmented Generation for Pathology VLMs via Reinforcement Learning (AAAI 2026 · similarity 0.85)
- PromptMRG: Diagnosis-Driven Prompts for Medical Report Generation (AAAI 2024 · similarity 0.85)
- Bootstrapping Large Language Models for Radiology Report Generation (AAAI 2024 · similarity 0.84)
- DAMPER: A Dual-Stage Medical Report Generation Framework with Coarse-Grained MeSH Alignment and Fine-Grained Hypergraph Matching (AAAI 2025 · similarity 0.84)
- DART: Disease-aware Image-Text Alignment and Self-correcting Re-alignment for Trustworthy Radiology Report Generation (CVPR 2025 · similarity 0.84)
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.