List your company

Image Captioning on MSCOCO (test)

149.6CIDErOFA
-3.737636.071275.88115.6888May 4, 2022Dec 15, 2022Jul 28, 2023Mar 10, 2024Oct 21, 2024Jun 3, 2025Jan 15, 2026

Evaluation results

MethodLinks
OFA2022.05149.643.531.926.1—————
LEMONoptimization=SCST2022.05145.542.631.425.5—————
OFA*Condition=Original score (no attack)2023.10145.4342.8131.325.37—————
CoCa2022.05143.640.933.924.7—————
SimVLM2022.05143.340.633.725.4—————
VinVL2022.05140.94131.125.4—————
LEMON2022.05139.141.530.824.1—————
BLIP2022.05136.740.4———————
CLIP-ViL2022.05134.240.229.723.8—————
BaselineSource Model=CLIP ViT-B/16, Target Model=BLIP2026.01131.3839.3130.6923.5259.62————
BaselineSource Model=ALBEF, Target Model=BLIP2026.01131.3839.3130.6923.5259.62————
AdvCLIPSource Model=CLIP ViT-B/16, Target Model=BLIP2026.01127.1438.230.0322.8858.7————
AdvCLIPSource Model=ALBEF, Target Model=BLIP2026.01121.7936.5929.1521.9557.35————
SGASource Model=CLIP ViT-B/16, Target Model=BLIP2026.01120.4736.128.8921.8557.25————
FD-UAPSource Model=CLIP ViT-B/16, Target Model=BLIP2026.01119.6335.9128.7821.6557.15————
C-PGCSource Model=CLIP ViT-B/16, Target Model=BLIP2026.01117.3335.6328.4421.3856.75————
ETUSource Model=CLIP ViT-B/16, Target Model=BLIP2026.01114.8134.8328.0721.0256.29————
FD-UAPSource Model=ALBEF, Target Model=BLIP2026.01113.2634.3728.0920.8656.02————
C-PGCSource Model=ALBEF, Target Model=BLIP2026.01112.6933.9727.7320.555.5————
SGASource Model=ALBEF, Target Model=BLIP2026.01112.2833.9227.6720.6255.41————
ETUSource Model=ALBEF, Target Model=BLIP2026.01109.3433.4227.2820.1154.96————
HRASource Model=ALBEF, Target Model=BLIP2026.01108.1932.6727.2120.1554.62————
HRASource Model=CLIP ViT-B/16, Target Model=BLIP2026.01106.632.4126.819.6154.31————
DRTarget Model=OFA2023.1095.5230.2624.4717.89—————
FDATarget Model=OFA2023.1055.7517.7717.9211.36—————
SSPTarget Model=OFA2023.1023.5410.9912.525.67—————
InfoNCE+LM+CSTraining Strategy=Blip2 stage one, Alignment Component=CS-Aligner2025.0216.7———————41.3
InfoNCE+LMTraining Strategy=Blip2 stage one2025.0214.3———————40.4
BSATarget Model=OFA2023.102.163.048.081.5—————
COATarget Model=Qwen2.5-VL-7B-Instruct, epsilon_img=8/2552025.12—————93.590.8821.47—
COATarget Model=InternVL3.5-4B, epsilon_img=8/2552025.12—————95.380.9261.53—
COATarget Model=LLaVA-1.5-7B, epsilon_img=8/2552025.12—————94.740.9170—
EGATarget Model=Qwen2.5-VL-7B-Instruct, epsilon_img=8/2552025.12—————94.810.88342.53—
EGATarget Model=InternVL3.5-4B, epsilon_img=8/2552025.12—————93.750.86237.29—
EGATarget Model=LLaVA-1.5-7B, epsilon_img=8/2552025.12—————93.120.84647.05—
MIETarget Model=Qwen2.5-VL-7B-Instruct, epsilon_img=8/2552025.12—————94.180.89213.95—
MIETarget Model=InternVL3.5-4B, epsilon_img=8/2552025.12—————94.830.90523.05—
MIETarget Model=LLaVA-1.5-7B, epsilon_img=8/2552025.12—————93.590.89321.29—
PGDTarget Model=Qwen2.5-VL-7B-Instruct, epsilon_img=8/2552025.12—————91.160.8421.28—
PGDTarget Model=InternVL3.5-4B, epsilon_img=8/2552025.12—————88.980.7931.41—
PGDTarget Model=LLaVA-1.5-7B, epsilon_img=8/2552025.12—————90.750.8341.52—
VLATarget Model=Qwen2.5-VL-7B-Instruct, epsilon_img=8/2552025.12—————89.220.8011.49—
VLATarget Model=InternVL3.5-4B, epsilon_img=8/2552025.12—————89.490.8041.4—
VLATarget Model=LLaVA-1.5-7B, epsilon_img=8/2552025.12—————87.750.7780—