PET/CT Report Impression Generation on PET F2I 41K
0.7075BLEU-4PET-F2I-7B
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| PET-F2I-7BCategory=Ours2026.03 | 0.7075 | 0.8673 | 0.8942 | 0.9103 | 0.9649 | 0.8074 | 0.1649 | 0.942 | |
| Claude Opus 4.5Category=Proprietary2026.03 | 0.2532 | 0.352 | 0.5692 | 0.759 | 0.6433 | 0.5268 | 0.5417 | 0.9814 | |
| DeepSeek V3.2Category=Open-Source (Large)2026.03 | 0.2401 | 0.3822 | 0.5296 | 0.7566 | 0.6339 | 0.508 | 0.5704 | 0.9778 | |
| Claude Sonnet 4.5Category=Proprietary2026.03 | 0.2374 | 0.3587 | 0.5369 | 0.7562 | 0.6286 | 0.5222 | 0.5434 | 0.9734 | |
| Kimi-K2Category=Open-Source (Large)2026.03 | 0.219 | 0.3398 | 0.5052 | 0.7557 | 0.6152 | 0.4638 | 0.5062 | 0.9548 | |
| Gemini 2.5 ProCategory=Proprietary2026.03 | 0.2112 | 0.4193 | 0.5127 | 0.7643 | 0.6597 | 0.4379 | 0.5152 | 0.9618 | |
| GLM-4.6Category=Open-Source (Large)2026.03 | 0.2067 | 0.3639 | 0.4872 | 0.7529 | 0.6139 | 0.4381 | 0.5329 | 0.9714 | |
| Gemini 2.0 FlashCategory=Proprietary2026.03 | 0.1897 | 0.4792 | 0.4653 | 0.7516 | 0.6401 | 0.4153 | 0.4947 | 0.9628 | |
| GPT-4oCategory=Proprietary2026.03 | 0.1876 | 0.3248 | 0.4577 | 0.7344 | 0.599 | 0.4346 | 0.5883 | 0.9756 | |
| BioGPTCategory=Medical Domain2026.03 | 0.1761 | 0.3294 | 0.4323 | 0.7348 | 0.6061 | 0.3857 | 0.5609 | 0.947 | |
| BioMedLMCategory=Medical Domain2026.03 | 0.175 | 0.3305 | 0.4295 | 0.7365 | 0.6027 | 0.3834 | 0.5578 | 0.953 | |
| MeditronCategory=Medical Domain2026.03 | 0.1728 | 0.3299 | 0.4253 | 0.7364 | 0.6039 | 0.3879 | 0.5504 | 0.9482 | |
| Baichuan-M3Category=Medical Domain2026.03 | 0.1705 | 0.3359 | 0.4248 | 0.7359 | 0.608 | 0.3719 | 0.5491 | 0.94 | |
| Clinical CamelCategory=Medical Domain2026.03 | 0.1685 | 0.3227 | 0.4201 | 0.7346 | 0.6006 | 0.3631 | 0.5583 | 0.9404 | |
| GPT-5.1Category=Proprietary2026.03 | 0.1681 | 0.2963 | 0.4768 | 0.7327 | 0.5888 | 0.4622 | 0.625 | 0.9898 | |
| PMC-LLaMACategory=Medical Domain2026.03 | 0.1643 | 0.331 | 0.4187 | 0.7332 | 0.5964 | 0.3653 | 0.5542 | 0.936 | |
| MedGemma-4BCategory=Medical Domain2026.03 | 0.1626 | 0.3345 | 0.4133 | 0.7348 | 0.5967 | 0.3597 | 0.5481 | 0.9328 | |
| Med-PaLM 2Category=Medical Domain2026.03 | 0.1563 | 0.3222 | 0.4032 | 0.7298 | 0.5854 | 0.3544 | 0.5484 | 0.9268 | |
| Qwen3-8BCategory=Open-Source (7B)2026.03 | 0.1552 | 0.3365 | 0.384 | 0.7076 | 0.7457 | 0.3349 | 0.6464 | 0.9368 | |
| DeepSeek R1Category=Open-Source (Large)2026.03 | 0.1517 | 0.2334 | 0.3805 | 0.7003 | 0.5489 | 0.3686 | 0.612 | 0.9024 | |
| Qwen2.5-7BCategory=Open-Source (7B)2026.03 | 0.1495 | 0.3152 | 0.3739 | 0.7234 | 0.5854 | 0.3387 | 0.6567 | 0.9522 | |
| Qwen3-235BCategory=Open-Source (Large)2026.03 | 0.148 | 0.2827 | 0.4201 | 0.7322 | 0.6194 | 0.5612 | 0.6709 | 0.9702 | |
| MiniMax-M2Category=Open-Source (Large)2026.03 | 0.1332 | 0.2323 | 0.3588 | 0.6875 | 0.5697 | 0.3292 | 0.6681 | 0.859 | |
| InternLM2.5-7BCategory=Open-Source (7B)2026.03 | 0.128 | 0.2887 | 0.2618 | 0.7029 | 0.5854 | 0.3271 | 0.6833 | 0.9452 | |
| Seed-Rice-7BCategory=Open-Source (7B)2026.03 | 0.1225 | 0.2987 | 0.2371 | 0.701 | 0.6076 | 0.3477 | 0.7207 | 0.9727 | |
| Hunyuan-MT-7BCategory=Open-Source (7B)2026.03 | 0.1045 | 0.2603 | 0.2688 | 0.6889 | 0.586 | 0.2487 | 0.6836 | 0.8214 | |
| GLM-4-9BCategory=Open-Source (7B)2026.03 | 0.0733 | 0.2189 | 0.2865 | 0.6929 | 0.6286 | 0.211 | 0.5898 | 0.7178 |