Image Captioning Evaluation on Flickr8k Expert
60.8Kendall Tau-c (tau_c)Pomms
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PommsEvaluation Framework=Pomms, Ensemble Composition=(ii) + Eagle2-9B2026.04 | 60.8 | 62.7 | |
| RIGELReference Requirement=Reference-free2026.06 | 59.9 | 59.5 | |
| G-VEvalReference-based=false, Test time [hour]=112025.12 | 59.7 | 61.5 | |
| GPT-4oEvaluation Framework=G-VEval2026.04 | 59.7 | 61.5 | |
| PommsEvaluation Framework=Pomms, Ensemble Composition=(i) GPT-4o and InternVL2.5-8B2026.04 | 59.7 | 61.5 | |
| G-VEvalReference Requirement=Reference-free2026.06 | 59.7 | 61.5 | |
| RIGELReference Requirement=Reference-based2026.06 | 59 | 58.6 | |
| PommsEvaluation Framework=Pomms, Ensemble Composition=(iii) + LLaVA-OneVision-7B2026.04 | 58.8 | 60.6 | |
| Ref-G-VEvalReference-based=true, Test time [hour]=112025.12 | 58.7 | 60.5 | |
| G-VEval-C (GPT-4o)FF=true2025.12 | 58.7 | 60.5 | |
| G-VEvalReference Requirement=Reference-based2026.06 | 58.7 | 60.5 | |
| PearlReference-based=true, Backbone=ViT-B/32, Test time [hour]=0.0432025.12 | 58.6 | 58.2 | |
| PearlReference-based=true, Backbone=ViT-B/322025.12 | 58.6 | 58.2 | |
| PearlReference Requirement=Reference-based2026.06 | 58.6 | 58.2 | |
| HiFiScoreReference-based=false2025.12 | 58.4 | — | |
| Ref-HiFiScoreReference-based=true2025.12 | 58.4 | — | |
| HiFiScoreReference Requirement=Reference-based2026.06 | 58.4 | — | |
| HiFiScoreReference Requirement=Reference-free2026.06 | 58.4 | — | |
| DISCODEBackbone=InternVL, Reference Requirement=Reference-free2026.06 | 58.1 | 57.7 | |
| RefPAC-S++ViT-LFF=false2025.12 | 57.9 | — | |
| PommsEvaluation Framework=Pomms, Ensemble Composition=(v) + Qwen2.5-VL-7B2026.04 | 57.8 | 59.6 | |
| PommsEvaluation Framework=Pomms, Ensemble Composition=(vi) + Phi-3.5-Vision2026.04 | 57.8 | 59.6 | |
| Ref-HICEScoreReference-based=true2025.12 | 57.7 | 57.2 | |
| Ref-HICEScoreReference Requirement=Reference-based2026.06 | 57.7 | 57.2 | |
| RefDISCODE-INFF=true2025.12 | 57.5 | 57.1 | |
| PommsEvaluation Framework=Pomms, Ensemble Composition=(iv) + DeepSeek-VL22026.04 | 57.3 | 59 | |
| DENEB (ViT-L)FF=false2025.12 | 56.8 | — | |
| EXPERTReference-based=false2025.12 | 56.7 | — | |
| EXPERTReference Requirement=Reference-free2026.06 | 56.7 | — | |
| PearlReference-based=false, Backbone=ViT-B/32, Test time [hour]=0.0432025.12 | 56.6 | 56.2 | |
| PearlReference-based=false, Backbone=ViT-B/322025.12 | 56.6 | 56.2 | |
| PearlReference Requirement=Reference-free2026.06 | 56.6 | 56.2 | |
| DENEBReference-based=true, Backbone=ViT-B/322025.12 | 56.5 | 55.6 | |
| RefDISCODE-LVFF=true2025.12 | 56.5 | 56.1 | |
| DENEBReference Requirement=Reference-based2026.06 | 56.5 | 55.6 | |
| PolosReference-based=true2025.12 | 56.4 | 56.1 | |
| HICEScoreReference-based=false2025.12 | 56.4 | 55.9 | |
| PolosFF=false2025.12 | 56.4 | — | |
| PolosReference Requirement=Reference-based2026.06 | 56.4 | 56.1 | |
| HICEScoreReference Requirement=Reference-free2026.06 | 56.4 | 55.9 | |
| DISCODEBackbone=LLaVA, Reference Requirement=Reference-free2026.06 | 56.1 | 55.7 | |
| RefPAC-SReference-based=true, Backbone=ViT-B/322025.12 | 55.9 | 55.5 | |
| RefPAC-S2025.12 | 55.9 | 55.5 | |
| RefPAC-SReference Requirement=Reference-based2026.06 | 55.9 | 55.5 | |
| BRIDGEReference-based=false2025.12 | 55.8 | 55.4 | |
| BRIDGEReference Requirement=Reference-free2026.06 | 55.8 | 55.4 | |
| RefPAC-S++Reference-based=true, Backbone=ViT-B/322025.12 | 55.7 | 55.3 | |
| RefPAC-S++ViT-B2025.12 | 55.7 | 55.3 | |
| RefPAC-S++Reference Requirement=Reference-based2026.06 | 55.7 | 55.3 | |
| InfoMetIC+Type=w/o ref2023.05 | 55.5 | — | |
| MIDFF=true2025.12 | 54.9 | — | |
| PAC-S++Reference-based=false, Backbone=ViT-B/322025.12 | 54.5 | 54.1 | |
| PAC-S++Reference Requirement=Reference-free2026.06 | 54.5 | 54.1 | |
| CLIP-StuneType=w/o ref2023.05 | 54.3 | — | |
| PAC-SReference-based=false, Backbone=ViT-B/322025.12 | 54.3 | 53.9 | |
| PAC-SReference Requirement=Reference-free2026.06 | 54.3 | 53.9 | |
| InfoMetICType=w/o ref2023.05 | 54.2 | — | |
| Qwen2.5-VL-7BEvaluation Framework=G-VEval2026.04 | 54 | 54.6 | |
| RefCLIP-SType=w/ ref2023.05 | 53 | — | |
| RefCLIP-SReference-based=true, Backbone=ViT-B/322025.12 | 53 | 52.6 | |
| FLEURReference-based=false2025.12 | 53 | — | |
| RefCLIP-SFF=true2025.12 | 53 | 52.6 | |
| RefCLIP-SReference Requirement=Reference-based2026.06 | 53 | 52.6 | |
| FLEURReference Requirement=Reference-free2026.06 | 53 | — | |
| InternVL2.5-8BEvaluation Framework=G-VEval2026.04 | 52.9 | 54.6 | |
| FAIEr-4Type=w/ ref2023.05 | 52.6 | — | |
| BLIP2ScoreReference-based=false2025.12 | 52.5 | 52.2 | |
| BLIP2-ScoreReference Requirement=Reference-free2026.06 | 52.5 | 52.2 | |
| Ref-FLEURReference-based=true2025.12 | 51.9 | — | |
| RefFLEURFF=true2025.12 | 51.9 | — | |
| Ref-FLEURReference Requirement=Reference-based2026.06 | 51.9 | — | |
| CLIP-SType=w/o ref2023.05 | 51.5 | — | |
| CLIP-SReference-based=false2025.12 | 51.2 | 51.1 | |
| CLIP-SReference Requirement=Reference-free2026.06 | 51.2 | 51.1 | |
| ViLBERTScore-FType=w/ ref2023.05 | 50.1 | — | |
| FAIEr-rType=w/o ref2023.05 | 50.1 | — | |
| ViLBERTScoreFF=true2025.12 | 50.1 | — | |
| TIGErType=w/ ref2023.05 | 49.3 | — | |
| TIGErFF=true2025.12 | 49.3 | — | |
| CLAIRReference-based=true2025.12 | 48.8 | 58.3 | |
| CLAIRReference Requirement=Reference-based2026.06 | 48.8 | 58.3 | |
| CLAIRFF=true2025.12 | 48.3 | — | |
| SPARCSReference-based=true, Test time [hour]=< 0.012025.12 | 48.1 | 43.8 | |
| UMICType=w/o ref2023.05 | 46.8 | — | |
| BERT-S++Type=w/ ref2023.05 | 46.7 | — | |
| BERTScoreReference-based=true, Test time [hour]=0.0402025.12 | 46.7 | 37.8 | |
| BERT-S++FF=true2025.12 | 46.7 | — | |
| BERTScoreReference Requirement=Reference-based2026.06 | 46.7 | 37.8 | |
| SPICEType=w/ ref2023.05 | 44.9 | — | |
| SPICEReference-based=true2025.12 | 44.9 | 51.7 | |
| SPICEFF=false2025.12 | 44.9 | 51.7 | |
| SPICEReference Requirement=Reference-based2026.06 | 44.9 | 51.7 | |
| CIDErType=w/ ref2023.05 | 43.9 | — | |
| CIDErReference-based=true2025.12 | 43.9 | 43.6 | |
| CIDErFF=false2025.12 | 43.9 | 43.6 | |
| CIDErReference Requirement=Reference-based2026.06 | 43.9 | 43.6 | |
| METEORType=w/ ref2023.05 | 41.8 | — | |
| METEORReference-based=true2025.12 | 41.8 | 41.5 | |
| METEORFF=false2025.12 | 41.8 | 41.5 | |
| METEORReference Requirement=Reference-based2026.06 | 41.8 | 41.5 |