Radiology Report Generation on CT-RATE (test)
60BL-1ASAP
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ASAPBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 60 | 47.2 | 39.2 | 34.4 | — | — | — | — | — | — | — | — | — | — | 75.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HLIPBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 59.8 | 46.9 | 39 | 34.2 | — | — | — | — | — | — | — | — | — | — | 75.9 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PCRLv2Backbone=3D ViT-B [82], Method Category=3D Med-SSL2026.05 | 59 | 46.5 | 38.4 | 33.7 | — | — | — | — | — | — | — | — | — | — | 75.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| fVLMBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 58.6 | 46.2 | 38.9 | 34.1 | — | — | — | — | — | — | — | — | — | — | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MAEBackbone=3D ViT-B [82], Method Category=3D Med-SSL2026.05 | 58.4 | 45.9 | 38.2 | 33 | — | — | — | — | — | — | — | — | — | — | 75.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HEMAEBackbone=3D ViT-B [82], Method Category=3D Med-SSL2026.05 | 58.3 | 45.9 | 38 | 32.8 | — | — | — | — | — | — | — | — | — | — | 75.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| M3AEBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 58.3 | 45.9 | 38.6 | 33.5 | — | — | — | — | — | — | — | — | — | — | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| MRMBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 58.2 | 45.9 | 38.8 | 33.8 | — | — | — | — | — | — | — | — | — | — | 75.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SimCroPBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 58.2 | 45.9 | 38.5 | 33.9 | — | — | — | — | — | — | — | — | — | — | 75.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-CLIPBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 58.1 | 45.7 | 38.5 | 33.9 | — | — | — | — | — | — | — | — | — | — | 75.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| M3DBackbone=3D ViT-B [82], Method Category=3D Med-VLP2026.05 | 57.8 | 46.3 | 38.9 | 33.8 | — | — | — | — | — | — | — | — | — | — | 75.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Random initBackbone=3D ViT-B [82], Method Category=Baseline2026.05 | 56.5 | 44.1 | 37.1 | 32.2 | — | — | — | — | — | — | — | — | — | — | 74.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RadFMCondition=Fine-tuned2026.03 | 50.2 | 33.88 | 24.09 | 17.02 | — | — | — | — | — | — | 86.17 | — | — | 30.46 | — | 36.81 | 36.1 | 13.48 | 19.63 | 28.51 | 10.16 | 13.05 | 36.1 | 13.48 | 17.74 | 20.47 | 10.11 | 12.14 | — | — | — | — | — | — | — | — | — | |
| InternVL3-AbSteerCondition=AbSteered2026.03 | 48.32 | 36.66 | 29.17 | 23.58 | — | — | — | — | — | — | 87.59 | — | — | 40.49 | — | 44.13 | 57.88 | 51.58 | 54.55 | 54.08 | 45.01 | 47.66 | 56.26 | 51.58 | 52.8 | 51.53 | 45.13 | 44.8 | — | — | — | — | — | — | — | — | — | |
| CT2RepCondition=Fine-tuned2026.03 | 47.91 | 38.23 | 32.41 | 28.04 | — | — | — | — | — | — | 88.1 | — | — | 45.43 | — | 45.62 | 26.39 | 10.5 | 14.1 | 17.03 | 12.59 | 10.65 | 24.44 | 10.37 | 11.35 | 17.36 | 9.07 | 10.86 | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-VL-AbSteerCondition=AbSteered2026.03 | 45.64 | 34.14 | 26.81 | 21.4 | — | — | — | — | — | — | 87.13 | — | — | 37.99 | — | 41.61 | 49.15 | 43.22 | 45.99 | 43.13 | 35.78 | 37.9 | 47.05 | 43.22 | 44.05 | 44.86 | 37.58 | 37.39 | — | — | — | — | — | — | — | — | — | |
| InternVL3-8BCondition=Fine-tuned2026.03 | 45.57 | 34.36 | 27.31 | 22.05 | — | — | — | — | — | — | 87.4 | — | — | 38.49 | — | 41.82 | 53.57 | 37.99 | 44.45 | 48.28 | 33.71 | 38.91 | 52.35 | 37.99 | 43.28 | 40 | 31.16 | 32.14 | — | — | — | — | — | — | — | — | — | |
| M3D-AbSteerCondition=AbSteered2026.03 | 45.22 | 34.64 | 27.99 | 23.09 | — | — | — | — | — | — | 87.83 | — | — | 38.58 | — | 41.26 | 44.95 | 41.66 | 43.24 | 39.94 | 34.75 | 36.18 | 44.01 | 41.66 | 41.89 | 43.19 | 37.26 | 36.54 | — | — | — | — | — | — | — | — | — | |
| M3D-8BCondition=Fine-tuned2026.03 | 44.95 | 34.36 | 27.83 | 22.98 | — | — | — | — | — | — | 87.52 | — | — | 37.76 | — | 40.75 | 47.6 | 28.54 | 35.69 | 41.62 | 22.41 | 26.74 | 47.06 | 28.54 | 33.13 | 35.04 | 23.21 | 25.21 | — | — | — | — | — | — | — | — | — | |
| Reg2RGCondition=Fine-tuned2026.03 | 44.89 | 33.43 | 26.42 | 21.08 | — | — | — | — | — | — | 86.18 | — | — | 24.41 | — | 38.25 | 28.47 | 11.06 | 15.93 | 19.76 | 8.17 | 10.48 | 26.19 | 11.06 | 14.51 | 19.81 | 11.51 | 12.19 | — | — | — | — | — | — | — | — | — | |
| Qwen2.5-VL-7BCondition=Fine-tuned2026.03 | 43.67 | 32.91 | 26.2 | 21.25 | — | — | — | — | — | — | 87.3 | — | — | 36.71 | — | 40.73 | 48.06 | 25.88 | 33.64 | 39.85 | 20.06 | 25.57 | 45.98 | 25.88 | 32.19 | 35.71 | 22.36 | 24.95 | — | — | — | — | — | — | — | — | — | |
| CT-CHATDimension=3D2026.01 | 43.64 | — | — | — | 23.06 | 31.57 | — | — | — | 0.221 | 88.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-CTCHATCondition=Fine-tuned2026.03 | 42.81 | 31.07 | 23.38 | 17.63 | — | — | — | — | — | — | 86.35 | — | — | 32.5 | — | 37.62 | 25.13 | 37.48 | 30.08 | 19.97 | 29.68 | 21.66 | 25.7 | 37.48 | 28.35 | 25.25 | 34.35 | 25.31 | — | — | — | — | — | — | — | — | — | |
| MedVL-SAM2Dimension=3D2026.01 | 41.85 | — | — | — | 22.15 | 34.59 | — | — | — | 0.237 | 89.33 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LLaVADimension=2D2026.01 | 18.47 | — | — | — | 11.22 | 10.34 | — | — | — | 0.05 | 30.55 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| M3DDimension=3D2026.01 | 14.49 | — | — | — | 14.11 | 19.25 | — | — | — | 0.081 | 84.12 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CXR-LLaVADimension=2D2026.01 | 9.12 | — | — | — | 2.86 | 5.05 | — | — | — | 0.049 | 39.18 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| LLaVA-MedDimension=2D2026.01 | 9.01 | — | — | — | 2.24 | 5.24 | — | — | — | 0.006 | 40.65 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-BERT2026.03 | 0.486 | — | — | 0.254 | — | — | 32.1 | 35.4 | 31 | — | — | — | — | 35.1 | — | 0.275 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ours-LLaMA2026.03 | 0.481 | — | — | 0.225 | — | — | 35.4 | 32.8 | 30.8 | — | — | — | — | 31.8 | — | 0.266 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT2Replongitudinal_data=false2024.03 | 0.46 | 0.415 | 0.39 | 0.369 | 0.295 | 0.459 | 74.9 | 54.8 | 53.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| M2KT2026.03 | 0.456 | — | — | 0.261 | — | — | 32.3 | 12 | 15.6 | — | — | — | — | 34.8 | — | 0.255 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RadFM2026.03 | 0.456 | — | — | 0.259 | — | — | 14.3 | 15.4 | 16.6 | — | — | — | — | 38.2 | — | 0.278 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-CLIP2026.03 | 0.455 | — | — | 0.259 | — | — | 21.7 | 12.9 | 14.5 | — | — | — | — | 38.3 | — | 0.277 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| GLoRIA2026.03 | 0.448 | — | — | 0.22 | — | — | 29.8 | 25.1 | 24 | — | — | — | — | 34.6 | — | 0.27 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT2Rep2026.03 | 0.447 | — | — | 0.251 | — | — | 20.6 | 7.3 | 9.9 | — | — | — | — | 37.7 | — | 0.275 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Base w/ CT-Netlongitudinal_data=false, backbone=CT-Net2024.03 | 0.443 | 0.399 | 0.375 | 0.354 | 0.286 | 0.442 | 51.3 | 53.1 | 45.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| PromptMRG2026.03 | 0.434 | — | — | 0.186 | — | — | 35.3 | 28.6 | 28.8 | — | — | — | — | 36.1 | — | 0.207 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| R2GenCMN2026.03 | 0.431 | — | — | 0.243 | — | — | 18.7 | 9.4 | 9.1 | — | — | — | — | 35 | — | 0.266 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| R2Gen2026.03 | 0.426 | — | — | 0.241 | — | — | 14 | 2.6 | 4.3 | — | — | — | — | 33.3 | — | 0.248 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT2RepLonglongitudinal_data=true2024.03 | 0.374 | 0.327 | 0.304 | 0.281 | 0.285 | 0.363 | 72.7 | 51.1 | 53.6 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Baseline (Longitudinal)longitudinal_data=true2024.03 | 0.372 | 0.317 | 0.282 | 0.251 | 0.238 | 0.353 | 66.6 | 46.5 | 52.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Baseline (Longitudinal) + report + volumelongitudinal_data=true, input_components=prior report + volume2024.03 | 0.365 | 0.319 | 0.292 | 0.271 | 0.239 | 0.351 | 65.8 | 41 | 53.3 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Baseline (Longitudinal) + reportlongitudinal_data=true, input_components=prior report2024.03 | 0.33 | 0.284 | 0.26 | 0.241 | 0.213 | 0.313 | 62.3 | 41 | 52.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-SSGEncoder=CT-SSG2025.10 | 0.317 | — | — | — | — | — | — | — | 38.72 | — | — | 0.164 | 43.66 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Baseline (Longitudinal) + volumelongitudinal_data=true, input_components=prior volume2024.03 | 0.305 | 0.261 | 0.238 | 0.22 | 0.204 | 0.291 | 66.2 | 43.4 | 53 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ViViTEncoder=ViViT2025.10 | 0.29 | — | — | — | — | — | — | — | 27.54 | — | — | 0.15 | 40.01 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| ResNet3DEncoder=ResNet3D2025.10 | 0.286 | — | — | — | — | — | — | — | 25.45 | — | — | 0.148 | 39.87 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| R2GenGPT2026.03 | 0.166 | — | — | 0.091 | — | — | 32.6 | 20.5 | 22.8 | — | — | — | — | 16.1 | — | 0.116 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CLarGen2026.05 | — | — | — | — | — | — | 40.5 | 64.6 | 48.6 | — | — | — | 0.472 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CLarGen2026.05 | — | — | — | — | — | — | 40.5 | 64.6 | 48.6 | — | — | 0.335 | 0.472 | 29.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 20.8 | 0.011 | 279 | 0.054 | 1,826 | 0.73 | 0.628 | 0.622 | -0.006 | |
| CT-AGRGVisual Extractor=CT-ViT2024.08 | — | — | — | — | — | — | 37.75 | 55.35 | 42.13 | — | 0.864 | 0.19 | — | 27 | -2.936 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-AGRGVisual Extractor=CT-Net, Multi-label classification=true, Multi-task classification=true, Multi-Abnormality Embedding=true, Language model=true2024.08 | — | — | — | 0.172 | — | — | 45.74 | 62.59 | 50.11 | — | 0.867 | 0.196 | — | 28 | -2.714 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-AGRG (Ablation)Visual Extractor=CT-Net, Multi-label classification=true, Multi-task classification=false, Multi-Abnormality Embedding=false, Language model=true2024.08 | — | — | — | 0.083 | — | — | — | 19.05 | 23.8 | — | — | 0.127 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-AGRG (Ablation)Visual Extractor=CT-Net, Multi-label classification=true, Multi-task classification=true, Multi-Abnormality Embedding=false, Language model=true2024.08 | — | — | — | 0.154 | — | — | — | 56.56 | 46.11 | — | — | 0.186 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-AGRG (Ablation)Visual Extractor=CT-Net, Multi-label classification=true, Multi-task classification=false, Multi-Abnormality Embedding=true, Language model=true2024.08 | — | — | — | 0.146 | — | — | — | 58.89 | 47.22 | — | — | 0.189 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-CHAT2026.05 | — | — | — | — | — | — | 49.3 | 16.4 | 18.9 | — | — | — | 0.368 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CT-CHAT2026.05 | — | — | — | — | — | — | 49.3 | 16.4 | 18.9 | — | — | 0.399 | 0.368 | 38.2 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 27.7 | 0.002 | 142 | 0.006 | 392 | 0.877 | 0.035 | 0.207 | 0.173 | |
| CT2Rep2024.08 | — | — | — | — | — | — | 41.63 | 38.12 | 36.73 | — | 0.865 | 0.173 | — | 24.3 | -3.314 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ground truth2026.05 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 0.016 | 296 | 0.094 | 2,402 | — | 1 | 0.999 | -0.001 | |
| HULU-Med2026.05 | — | — | — | — | — | — | 30.3 | 16.6 | 19.4 | — | — | — | 0.365 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| HULU-Med2026.05 | — | — | — | — | — | — | 30.3 | 16.6 | 19.4 | — | — | 0.374 | 0.365 | 34 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 23.9 | 0.014 | 230 | 0.08 | 1,393 | 0.845 | 0.058 | 0.3 | 0.241 | |
| Reg2Rg2026.05 | — | — | — | — | — | — | 37.3 | 13.8 | 18.5 | — | — | — | 0.363 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Reg2Rg2026.05 | — | — | — | — | — | — | 37.3 | 13.8 | 18.5 | — | — | 0.368 | 0.363 | 24.5 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 27.5 | 0.004 | 151 | 0.018 | 528 | 0.923 | 0.088 | 0.195 | 0.107 | |
| SAMF2026.05 | — | — | — | — | — | — | 14.4 | 8.6 | 9.7 | — | — | — | 0.355 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SAMF2026.05 | — | — | — | — | — | — | 14.4 | 8.6 | 9.7 | — | — | 0.388 | 0.355 | 38.8 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | 26.9 | 0.001 | 114 | 0.002 | 271 | 0.918 | 0 | 0.187 | 0.187 |