Retrieval on MS-COCO
5.109Ave AesQCQC (Ours)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| QCQC (Ours)Aes Cond=H, Rel Cond=H, Controllable=true2026.02 | 5.109 | 0.391 | |
| GPT-4oAes Cond=H, Rel Cond=M, Controllable=false2026.02 | 5.075 | 0.352 | |
| FT-Blip2Aes Cond=H, Rel Cond=H, Controllable=false2026.02 | 5.053 | 0.349 | |
| GPT-4oAes Cond=H, Rel Cond=H, Controllable=false2026.02 | 5.048 | 0.351 | |
| FT-Blip2Aes Cond=H, Rel Cond=L, Controllable=false2026.02 | 5.039 | 0.343 | |
| QCQC (Ours)Aes Cond=H, Rel Cond=M, Controllable=true2026.02 | 5.02 | 0.368 | |
| QCQC (Ours)Aes Cond=H, Rel Cond=L, Controllable=true2026.02 | 5.017 | 0.353 | |
| FT-CoCaAes Cond=H, Rel Cond=M, Controllable=false2026.02 | 5.004 | 0.349 | |
| QCQC (Ours)Aes Cond=M, Rel Cond=L, Controllable=true2026.02 | 4.973 | 0.356 | |
| FT-Blip2Aes Cond=M, Rel Cond=M, Controllable=false2026.02 | 4.968 | 0.349 | |
| FT-Blip2Aes Cond=H, Rel Cond=M, Controllable=false2026.02 | 4.967 | 0.349 | |
| FTAes Cond=H, Rel Cond=H, Controllable=false2026.02 | 4.96 | 0.372 | |
| FT-CoCaAes Cond=H, Rel Cond=L, Controllable=false2026.02 | 4.95 | 0.347 | |
| FT-CoCaAes Cond=M, Rel Cond=M, Controllable=false2026.02 | 4.939 | 0.349 | |
| FTAes Cond=H, Rel Cond=M, Controllable=false2026.02 | 4.919 | 0.376 | |
| LLaMA3Aes Cond=M, Rel Cond=L, Controllable=false2026.02 | 4.916 | 0.348 | |
| QCQC (Ours)Aes Cond=M, Rel Cond=H, Controllable=true2026.02 | 4.916 | 0.387 | |
| FT-Blip2Aes Cond=M, Rel Cond=L, Controllable=false2026.02 | 4.914 | 0.344 | |
| LLaMA3Aes Cond=H, Rel Cond=H, Controllable=false2026.02 | 4.911 | 0.344 | |
| LLaMA3Aes Cond=L, Rel Cond=L, Controllable=false2026.02 | 4.903 | 0.348 | |
| FT-CoCaAes Cond=M, Rel Cond=H, Controllable=false2026.02 | 4.903 | 0.352 | |
| FT-CoCaAes Cond=H, Rel Cond=H, Controllable=false2026.02 | 4.898 | 0.351 | |
| QCQC (Ours)Aes Cond=L, Rel Cond=L, Controllable=true2026.02 | 4.896 | 0.354 | |
| LLaMA3Aes Cond=L, Rel Cond=M, Controllable=false2026.02 | 4.891 | 0.349 | |
| FT-CoCaAes Cond=L, Rel Cond=L, Controllable=false2026.02 | 4.89 | 0.347 | |
| FT-CoCaAes Cond=L, Rel Cond=M, Controllable=false2026.02 | 4.889 | 0.348 | |
| FTAes Cond=M, Rel Cond=H, Controllable=false2026.02 | 4.888 | 0.373 | |
| FT-CoCaAes Cond=M, Rel Cond=L, Controllable=false2026.02 | 4.885 | 0.346 | |
| FT-Blip2Aes Cond=L, Rel Cond=M, Controllable=false2026.02 | 4.883 | 0.351 | |
| LLaMA3Aes Cond=M, Rel Cond=H, Controllable=false2026.02 | 4.88 | 0.347 | |
| GPT-4oAes Cond=M, Rel Cond=H, Controllable=false2026.02 | 4.88 | 0.35 | |
| QCQC (Ours)Aes Cond=M, Rel Cond=M, Controllable=true2026.02 | 4.879 | 0.368 | |
| LLaMA3Aes Cond=M, Rel Cond=M, Controllable=false2026.02 | 4.875 | 0.349 | |
| FT-Blip2Aes Cond=M, Rel Cond=H, Controllable=false2026.02 | 4.873 | 0.35 | |
| LLaMA3Aes Cond=H, Rel Cond=L, Controllable=false2026.02 | 4.871 | 0.348 | |
| FTAes Cond=L, Rel Cond=H, Controllable=false2026.02 | 4.866 | 0.373 | |
| LLaMA3Aes Cond=H, Rel Cond=M, Controllable=false2026.02 | 4.858 | 0.35 | |
| LLaMA3Aes Cond=L, Rel Cond=H, Controllable=false2026.02 | 4.855 | 0.347 | |
| PTAes Cond=L, Rel Cond=H, Controllable=false2026.02 | 4.855 | 0.35 | |
| PTAes Cond=M, Rel Cond=H, Controllable=false2026.02 | 4.854 | 0.345 | |
| FTAes Cond=M, Rel Cond=M, Controllable=false2026.02 | 4.852 | 0.369 | |
| PTAes Cond=H, Rel Cond=H, Controllable=false2026.02 | 4.847 | 0.344 | |
| GPT-4oAes Cond=H, Rel Cond=L, Controllable=false2026.02 | 4.838 | 0.359 | |
| FTAes Cond=H, Rel Cond=L, Controllable=false2026.02 | 4.833 | 0.367 | |
| PTAes Cond=H, Rel Cond=L, Controllable=false2026.02 | 4.83 | 0.351 | |
| FT-Blip2Aes Cond=L, Rel Cond=H, Controllable=false2026.02 | 4.824 | 0.352 | |
| PTAes Cond=M, Rel Cond=L, Controllable=false2026.02 | 4.821 | 0.349 | |
| FTAes Cond=L, Rel Cond=M, Controllable=false2026.02 | 4.82 | 0.369 | |
| PrefixAes Cond=L, Rel Cond=L, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=L, Rel Cond=M, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=L, Rel Cond=H, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=M, Rel Cond=L, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=M, Rel Cond=M, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=M, Rel Cond=H, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=H, Rel Cond=L, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=H, Rel Cond=M, Controllable=false2026.02 | 4.817 | 0.349 | |
| PrefixAes Cond=H, Rel Cond=H, Controllable=false2026.02 | 4.817 | 0.349 | |
| FTAes Cond=M, Rel Cond=L, Controllable=false2026.02 | 4.813 | 0.373 | |
| QCQC (Ours)Aes Cond=L, Rel Cond=M, Controllable=true2026.02 | 4.809 | 0.365 | |
| GPT-4oAes Cond=M, Rel Cond=M, Controllable=false2026.02 | 4.808 | 0.358 | |
| FT-CoCaAes Cond=L, Rel Cond=H, Controllable=false2026.02 | 4.793 | 0.356 | |
| FTAes Cond=L, Rel Cond=L, Controllable=false2026.02 | 4.785 | 0.369 | |
| GPT-4oAes Cond=M, Rel Cond=L, Controllable=false2026.02 | 4.782 | 0.36 | |
| FT-Blip2Aes Cond=L, Rel Cond=L, Controllable=false2026.02 | 4.776 | 0.349 | |
| PTAes Cond=M, Rel Cond=M, Controllable=false2026.02 | 4.775 | 0.344 | |
| GPT-4oAes Cond=L, Rel Cond=M, Controllable=false2026.02 | 4.754 | 0.357 | |
| PTAes Cond=L, Rel Cond=L, Controllable=false2026.02 | 4.742 | 0.347 | |
| PTAes Cond=L, Rel Cond=M, Controllable=false2026.02 | 4.731 | 0.345 | |
| PTAes Cond=H, Rel Cond=M, Controllable=false2026.02 | 4.726 | 0.347 | |
| QCQC (Ours)Aes Cond=L, Rel Cond=H, Controllable=true2026.02 | 4.719 | 0.385 | |
| GPT-4oAes Cond=L, Rel Cond=H, Controllable=false2026.02 | 4.686 | 0.354 | |
| GPT-4oAes Cond=L, Rel Cond=L, Controllable=false2026.02 | 4.673 | 0.371 |