Image Captioning on COCO-LN 500 (test)
49.06CAPTURE8B-Ours
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 8B-OursBase Model=InternVL2, Method=CIM2026.03 | 49.06 | 41.89 | 25.85 | 82.05 | 69.4 | 74.57 | 73.5 | 52.59 | 57.49 | 35.54 | |
| CIMBase Model=InternVL2-8B2026.03 | 49.06 | — | — | 82.05 | 69.4 | 74.57 | 73.5 | 52.59 | 57.49 | 35.54 | |
| 7B-OursBase Model=Qwen2.5-VL, Method=CIM2026.03 | 48.93 | 37.15 | 26.81 | 79.28 | 75.91 | 77.59 | 72.49 | 54.46 | 58.51 | 44.15 | |
| CIMBase Model=Qwen2.5-VL-7B2026.03 | 48.93 | — | — | 80.31 | 75.91 | 77.59 | 72.49 | 54.46 | 58.51 | 44.15 | |
| 8B-OursBase Model=InternVL3, Method=CIM2026.03 | 48.9 | 40.64 | 25.64 | 80.33 | 73.33 | 76.14 | 73.33 | 54.54 | 58.7 | 38.67 | |
| CIMBase Model=InternVL3-8B2026.03 | 48.9 | — | — | 80.33 | 73.33 | 76.14 | 73.33 | 54.54 | 58.7 | 38.67 | |
| 7B-OursBase Model=Qwen2-VL, Method=CIM2026.03 | 48.64 | 39.63 | 27.06 | 80.9 | 72.23 | 75.8 | 72.45 | 54.08 | 58.22 | 38.71 | |
| CIMBase Model=Qwen2-VL-7B2026.03 | 48.64 | — | — | 80.9 | 72.23 | 75.8 | 72.45 | 54.08 | 58.22 | 38.71 | |
| CIMBase Model=LLaVA1.5-7B2026.03 | 48.62 | — | — | 83.84 | 60.7 | 69.8 | 71.24 | 48.91 | 54.38 | 24.98 | |
| 78B-BaseBase Model=InternVL32026.03 | 48.48 | 40.6 | 19.5 | 82.46 | 65.02 | 72.07 | 71.76 | 46.22 | 52.5 | 29.61 | |
| 8B-OursBase Model=InternVL2.5, Method=CIM2026.03 | 48.28 | 39.27 | 22.85 | 80.31 | 70.42 | 74.53 | 71.85 | 51.71 | 56.07 | 36.96 | |
| CIMBase Model=InternVL2.5-8B2026.03 | 48.28 | — | — | 80.31 | 70.42 | 74.53 | 71.85 | 51.71 | 56.07 | 36.96 | |
| 78B-BaseBase Model=InternVL2.52026.03 | 48.25 | 41.22 | 20.79 | 81.38 | 64.59 | 71.42 | 71.25 | 46.84 | 52.85 | 28.92 | |
| 72B-BaseBase Model=Qwen2.5-VL2026.03 | 48.06 | 40.21 | 22.57 | 81.21 | 66.37 | 72.42 | 68.46 | 48.21 | 52.97 | 32.13 | |
| 8B-BaseBase Model=InternVL32026.03 | 47.88 | 33.21 | 16.16 | 82.51 | 63.31 | 71 | 71.14 | 43.97 | 50.66 | 26.44 | |
| BaseBase Model=InternVL3-8B2026.03 | 47.88 | — | — | 82.51 | 63.31 | 71 | 71.14 | 43.97 | 50.66 | 26.44 | |
| 76B-BaseBase Model=InternVL22026.03 | 47.64 | 37.31 | 22.29 | 80.57 | 67.25 | 72.66 | 70.69 | 46.9 | 54.5 | 31.6 | |
| 72B-BaseBase Model=Qwen2-VL2026.03 | 47.2 | 39.37 | 19.77 | 82.52 | 62.89 | 70.72 | 69.36 | 42.78 | 49.38 | 27.46 | |
| 8B-BaseBase Model=InternVL2.52026.03 | 47.09 | 38.04 | 18.54 | 82.42 | 64.02 | 71.42 | 72 | 46.03 | 52.5 | 27.29 | |
| BaseBase Model=InternVL2.5-8B2026.03 | 47.09 | — | — | 82.42 | 64.02 | 71.42 | 72 | 46.03 | 52.5 | 27.29 | |
| 7B-BaseBase Model=Qwen2-VL2026.03 | 46.52 | 39.57 | 20.42 | 81.12 | 61.82 | 69.47 | 66.48 | 42.86 | 48.68 | 20.47 | |
| BaseBase Model=Qwen2-VL-7B2026.03 | 46.52 | — | — | 81.12 | 61.82 | 69.47 | 66.48 | 42.86 | 48.68 | 20.47 | |
| 8B-BaseBase Model=InternVL22026.03 | 45.86 | 31.44 | 21.28 | 80.46 | 65.2 | 71.05 | 70.63 | 47.4 | 52.73 | 26.65 | |
| BaseBase Model=InternVL2-8B2026.03 | 45.86 | — | — | 80.46 | 65.2 | 71.05 | 70.63 | 47.4 | 52.73 | 26.65 | |
| BaseBase Model=LLaVA1.5-7B2026.03 | 44.75 | — | — | 81.2 | 59.02 | 67.56 | 58.61 | 37.2 | 42.34 | 14.38 | |
| 7B-BaseBase Model=Qwen2.5-VL2026.03 | 44.12 | 29.11 | 14.58 | 82.35 | 55.72 | 65.37 | 66.3 | 39.9 | 46.25 | 23.76 | |
| BaseBase Model=Qwen2.5-VL-7B2026.03 | 44.12 | — | — | 82.35 | 55.72 | 65.37 | 66.3 | 39.9 | 46.25 | 23.76 |