Chest X-ray Report Generation on IU Chest X-rays (test)
0.055BLEUR4Agent-Gemini
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| R4Agent-GeminiSetting=CoT / Agent2026.01 | 0.055 | 0.892 | 0.2229 | — | 8.21 | 7.27 | 9.1 | 8.02 | 7.51 | |
| LLaVA-Med-7BSetting=3-Shot2026.01 | 0.0549 | 0.8812 | 0.2272 | — | — | — | — | — | — | |
| Gemini-2.5-FlashSetting=3-Shot2026.01 | 0.0543 | 0.8758 | 0.2392 | — | — | — | — | — | — | |
| MedGemma-4BSetting=3-Shot2026.01 | 0.0533 | 0.878 | 0.2337 | — | — | — | — | — | — | |
| Med-Llama3-8BSetting=3-Shot2026.01 | 0.052 | 0.8991 | 0.2378 | — | — | — | — | — | — | |
| Gemini-2.5-FlashSetting=CoT / Agent2026.01 | 0.0511 | 0.8759 | 0.2381 | — | 6.73 | 5.54 | 5.47 | 5.58 | 4.58 | |
| Med-Llama3-8BSetting=CoT / Agent2026.01 | 0.0491 | 0.8836 | 0.241 | — | 7.01 | 5.21 | 5.55 | 5.6 | 4.61 | |
| LLaMA-3.2-11BSetting=3-Shot2026.01 | 0.0491 | 0.8734 | 0.2425 | — | — | — | — | — | — | |
| DeepSeek-VL2-3BSetting=CoT / Agent2026.01 | 0.0491 | 0.8836 | 0.241 | — | 5.97 | 5.09 | 4.83 | 5.02 | 4.2 | |
| LLaVA-Med-7BSetting=CoT / Agent2026.01 | 0.0476 | 0.8799 | 0.2323 | — | 6.67 | 4.98 | 5.62 | 5.45 | 4.54 | |
| Qwen2.5-VL-7BSetting=3-Shot2026.01 | 0.0466 | 0.8375 | 0.2243 | — | — | — | — | — | — | |
| Yi et al.Setting=CoT / Agent2026.01 | 0.0466 | 0.8819 | 0.2471 | — | 8.16 | 6.74 | 8.26 | 7.38 | 6.36 | |
| Qwen2.5-VL-7BSetting=CoT / Agent2026.01 | 0.0459 | 0.8388 | 0.2367 | — | 6.68 | 4.85 | 5.04 | 5.17 | 4.11 | |
| R4Agent-Qwen2.5-VL-7BSetting=CoT / Agent2026.01 | 0.0437 | 0.8791 | 0.2114 | — | 8.14 | 6.86 | 8.31 | 7.44 | 6.44 | |
| Gemma-3-4BSetting=CoT / Agent2026.01 | 0.042 | 0.8676 | 0.2212 | — | 6.32 | 5.29 | 5.35 | 5.28 | 4.17 | |
| R4Agent-MedGemma-4BSetting=CoT / Agent2026.01 | 0.0414 | 0.8767 | 0.1985 | — | 7.99 | 6.53 | 8.26 | 7.2 | 6.03 | |
| MedGemma-4BSetting=CoT / Agent2026.01 | 0.0408 | 0.8807 | 0.2329 | — | 6.55 | 5.48 | 5.4 | 5.48 | 4.49 | |
| LLaMA-3.2-11BSetting=CoT / Agent2026.01 | 0.0407 | 0.8661 | 0.2386 | — | 6.93 | 4.7 | 5.04 | 5.25 | 4.33 | |
| Gemma-3-4BSetting=3-Shot2026.01 | 0.0398 | 0.8577 | 0.2066 | — | — | — | — | — | — | |
| DeepSeek-VL2-3BSetting=3-Shot2026.01 | 0.0297 | 0.7738 | 0.2378 | — | — | — | — | — | — | |
| Gemini-2.5-FlashSetting=Zero-Shot2026.01 | 0.0238 | 0.8823 | 0.1593 | — | — | — | — | — | — | |
| Qwen2.5-VL-7BSetting=Zero-Shot2026.01 | 0.0234 | 0.8251 | 0.2198 | — | — | — | — | — | — | |
| LLaMA-3.2-11BSetting=Zero-Shot2026.01 | 0.022 | 0.8656 | 0.2299 | — | — | — | — | — | — | |
| LLaVA-Med-7BSetting=Zero-Shot2026.01 | 0.0201 | 0.8714 | 0.2286 | — | — | — | — | — | — | |
| Med-Llama3-8BSetting=Zero-Shot2026.01 | 0.0199 | 0.885 | 0.2379 | — | — | — | — | — | — | |
| Gemma-3-4BSetting=Zero-Shot2026.01 | 0.0183 | 0.8485 | 0.1994 | — | — | — | — | — | — | |
| MedGemma-4BSetting=Zero-Shot2026.01 | 0.0176 | 0.8682 | 0.2036 | — | — | — | — | — | — | |
| DeepSeek-VL2-3BSetting=Zero-Shot2026.01 | 0.0175 | 0.7845 | 0.119 | — | — | — | — | — | — |