Image Captioning Evaluation on LongCap-Arena (test)
42.3FluencyITIScore
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ITIScoreZero-shot=true2026.04 | 42.3 | 55.8 | 46.5 | |
| ChatGPT-5Zero-shot=true2026.04 | 38.6 | 50.1 | 42.5 | |
| Gemini-3.0-FlashZero-shot=true2026.04 | 37.2 | 52.7 | 40.9 | |
| Gemini-2.5-FlashZero-shot=true2026.04 | 35 | 46.6 | 35.2 | |
| Qwen3-VL (7B)Zero-shot=true2026.04 | 30.7 | 40.1 | 29.7 | |
| InternVL3.5 (7B)Zero-shot=true2026.04 | 28.6 | 43 | 31.8 | |
| CLIP-SZero-shot=true2026.04 | 25 | 25.9 | 20.6 | |
| PAC-SZero-shot=true2026.04 | 23.3 | 26.2 | 21.7 | |
| ChatGPT-4oZero-shot=true2026.04 | 23.1 | 48.9 | 37.1 | |
| BLEUZero-shot=true2026.04 | 14.5 | 30.3 | 6.3 | |
| FLEURZero-shot=true2026.04 | 11.8 | 15 | 6.6 | |
| BERT-SZero-shot=true2026.04 | 7.3 | 18.6 | 8.4 | |
| ROUGEZero-shot=true2026.04 | 6.2 | 8.3 | 5.7 | |
| METEORZero-shot=true2026.04 | 4.6 | 11.2 | 8.1 | |
| CIDErZero-shot=true2026.04 | 3.2 | 5.5 | 5.1 |