List your company

Image Captioning Evaluation on Flickr8k Expert

60.8Kendall Tau-c (tau_c)Pomms
41.0446.1751.356.43May 10, 2023Nov 17, 2023May 26, 2024Dec 3, 2024Jun 12, 2025Dec 20, 2025Jun 29, 2026

Evaluation results

MethodLinks
PommsEvaluation Framework=Pomms, Ensemble Composition=(ii) + Eagle2-9B2026.0460.862.7
RIGELReference Requirement=Reference-free2026.0659.959.5
G-VEvalReference-based=false, Test time [hour]=112025.1259.761.5
GPT-4oEvaluation Framework=G-VEval2026.0459.761.5
PommsEvaluation Framework=Pomms, Ensemble Composition=(i) GPT-4o and InternVL2.5-8B2026.0459.761.5
G-VEvalReference Requirement=Reference-free2026.0659.761.5
RIGELReference Requirement=Reference-based2026.065958.6
PommsEvaluation Framework=Pomms, Ensemble Composition=(iii) + LLaVA-OneVision-7B2026.0458.860.6
Ref-G-VEvalReference-based=true, Test time [hour]=112025.1258.760.5
G-VEval-C (GPT-4o)FF=true2025.1258.760.5
G-VEvalReference Requirement=Reference-based2026.0658.760.5
PearlReference-based=true, Backbone=ViT-B/32, Test time [hour]=0.0432025.1258.658.2
PearlReference-based=true, Backbone=ViT-B/322025.1258.658.2
PearlReference Requirement=Reference-based2026.0658.658.2
HiFiScoreReference-based=false2025.1258.4—
Ref-HiFiScoreReference-based=true2025.1258.4—
HiFiScoreReference Requirement=Reference-based2026.0658.4—
HiFiScoreReference Requirement=Reference-free2026.0658.4—
DISCODEBackbone=InternVL, Reference Requirement=Reference-free2026.0658.157.7
RefPAC-S++ViT-LFF=false2025.1257.9—
PommsEvaluation Framework=Pomms, Ensemble Composition=(v) + Qwen2.5-VL-7B2026.0457.859.6
PommsEvaluation Framework=Pomms, Ensemble Composition=(vi) + Phi-3.5-Vision2026.0457.859.6
Ref-HICEScoreReference-based=true2025.1257.757.2
Ref-HICEScoreReference Requirement=Reference-based2026.0657.757.2
RefDISCODE-INFF=true2025.1257.557.1
PommsEvaluation Framework=Pomms, Ensemble Composition=(iv) + DeepSeek-VL22026.0457.359
DENEB (ViT-L)FF=false2025.1256.8—
EXPERTReference-based=false2025.1256.7—
EXPERTReference Requirement=Reference-free2026.0656.7—
PearlReference-based=false, Backbone=ViT-B/32, Test time [hour]=0.0432025.1256.656.2
PearlReference-based=false, Backbone=ViT-B/322025.1256.656.2
PearlReference Requirement=Reference-free2026.0656.656.2
DENEBReference-based=true, Backbone=ViT-B/322025.1256.555.6
RefDISCODE-LVFF=true2025.1256.556.1
DENEBReference Requirement=Reference-based2026.0656.555.6
PolosReference-based=true2025.1256.456.1
HICEScoreReference-based=false2025.1256.455.9
PolosFF=false2025.1256.4—
PolosReference Requirement=Reference-based2026.0656.456.1
HICEScoreReference Requirement=Reference-free2026.0656.455.9
DISCODEBackbone=LLaVA, Reference Requirement=Reference-free2026.0656.155.7
RefPAC-SReference-based=true, Backbone=ViT-B/322025.1255.955.5
RefPAC-S2025.1255.955.5
RefPAC-SReference Requirement=Reference-based2026.0655.955.5
BRIDGEReference-based=false2025.1255.855.4
BRIDGEReference Requirement=Reference-free2026.0655.855.4
RefPAC-S++Reference-based=true, Backbone=ViT-B/322025.1255.755.3
RefPAC-S++ViT-B2025.1255.755.3
RefPAC-S++Reference Requirement=Reference-based2026.0655.755.3
InfoMetIC+Type=w/o ref2023.0555.5—
MIDFF=true2025.1254.9—
PAC-S++Reference-based=false, Backbone=ViT-B/322025.1254.554.1
PAC-S++Reference Requirement=Reference-free2026.0654.554.1
CLIP-StuneType=w/o ref2023.0554.3—
PAC-SReference-based=false, Backbone=ViT-B/322025.1254.353.9
PAC-SReference Requirement=Reference-free2026.0654.353.9
InfoMetICType=w/o ref2023.0554.2—
Qwen2.5-VL-7BEvaluation Framework=G-VEval2026.045454.6
RefCLIP-SType=w/ ref2023.0553—
RefCLIP-SReference-based=true, Backbone=ViT-B/322025.125352.6
FLEURReference-based=false2025.1253—
RefCLIP-SFF=true2025.125352.6
RefCLIP-SReference Requirement=Reference-based2026.065352.6
FLEURReference Requirement=Reference-free2026.0653—
InternVL2.5-8BEvaluation Framework=G-VEval2026.0452.954.6
FAIEr-4Type=w/ ref2023.0552.6—
BLIP2ScoreReference-based=false2025.1252.552.2
BLIP2-ScoreReference Requirement=Reference-free2026.0652.552.2
Ref-FLEURReference-based=true2025.1251.9—
RefFLEURFF=true2025.1251.9—
Ref-FLEURReference Requirement=Reference-based2026.0651.9—
CLIP-SType=w/o ref2023.0551.5—
CLIP-SReference-based=false2025.1251.251.1
CLIP-SReference Requirement=Reference-free2026.0651.251.1
ViLBERTScore-FType=w/ ref2023.0550.1—
FAIEr-rType=w/o ref2023.0550.1—
ViLBERTScoreFF=true2025.1250.1—
TIGErType=w/ ref2023.0549.3—
TIGErFF=true2025.1249.3—
CLAIRReference-based=true2025.1248.858.3
CLAIRReference Requirement=Reference-based2026.0648.858.3
CLAIRFF=true2025.1248.3—
SPARCSReference-based=true, Test time [hour]=< 0.012025.1248.143.8
UMICType=w/o ref2023.0546.8—
BERT-S++Type=w/ ref2023.0546.7—
BERTScoreReference-based=true, Test time [hour]=0.0402025.1246.737.8
BERT-S++FF=true2025.1246.7—
BERTScoreReference Requirement=Reference-based2026.0646.737.8
SPICEType=w/ ref2023.0544.9—
SPICEReference-based=true2025.1244.951.7
SPICEFF=false2025.1244.951.7
SPICEReference Requirement=Reference-based2026.0644.951.7
CIDErType=w/ ref2023.0543.9—
CIDErReference-based=true2025.1243.943.6
CIDErFF=false2025.1243.943.6
CIDErReference Requirement=Reference-based2026.0643.943.6
METEORType=w/ ref2023.0541.8—
METEORReference-based=true2025.1241.841.5
METEORFF=false2025.1241.841.5
METEORReference Requirement=Reference-based2026.0641.841.5