List your company

Visual Dialog on VisDial 1.0 (val)

0.6951MRRGST (Student)
0.1240360.2722930.420550.568807Sep 11, 2019Oct 22, 2020Dec 3, 2021Jan 14, 2023Feb 25, 2024Apr 7, 2025May 20, 2026

Evaluation results

MethodLinks
GST (Student)Pre-train # Images=6.7M2022.050.6951————65.92
BLIPevaluation_protocol=discriminative2022.010.694156.4485.993.33.2—
BLIPPre-train # Images=129M2022.050.6941—————
VD-ViLBERT†evaluation_protocol=discriminative, pre-training=ViLBERT with additional VQA data2022.010.69155.8885.593.293.25—
UTCsetting=discriminative2022.050.685855.4885.3893.23.280.6322
VD-BERTevaluation_protocol=discriminative2022.010.674454.0283.9692.333.53—
VDBERTsetting=discriminative2022.050.674454.0283.9692.333.530.6322
VisDial-BERTsetting=discriminative2022.050.671753.4284.4192.623.410.6096
ReDANDecoder Type=Discriminative2019.110.642150.681.3990.264.050.5932
ReDANsetting=discriminative2022.050.642150.681.3990.264.050.5932
ReDANDecoder Type=Discriminative2026.050.642150.681.3990.264.050.5932
HCIAEDecoder Type=Discriminative2019.110.629648.9480.589.664.240.5775
HCIAEsetting=discriminative2022.050.629648.9480.589.664.240.5775
HCIAEDecoder Type=Discriminative2026.050.629648.9480.589.664.240.5775
CoAttDecoder Type=Discriminative2019.110.629148.8680.4189.834.210.5772
CoAttsetting=discriminative2022.050.629148.8680.4189.834.210.5772
CoAttDecoder Type=Discriminative2026.050.629148.8680.4189.834.210.5772
LTMIDecoder Type=Discriminative2019.110.623248.9478.6587.884.860.6272
LTMIsetting=discriminative2022.050.623248.9478.6587.884.860.6272
LTMIDecoder Type=Discriminative2026.050.623248.9478.6587.884.860.6272
ACE2019.090.624782.492.33.810.53
PDUN2019.090.6247.382.592.63.680.54
NMN2019.090.6147.578.188.84.40.54
MNDecoder Type=Discriminative2019.110.604246.0978.1488.054.630.5513
MNsetting=discriminative2022.050.604246.0978.1488.054.630.5513
MNDecoder Type=Discriminative2026.050.604246.0978.1488.054.630.5513
ECE2019.090.5843.675.485.35.360.49
MN2019.090.5642.47484.35.590.47
LF2019.090.5540.972.482.85.950.45
HRE2019.090.5439.970.481.56.410.45
V²DialSent. Trans.=OpenAI TE-3, trained_and_evaluated_by_authors=true2025.030.53246.159.365.7—0.531
Student2022.050.531943.0864.0971.5114.340.6547
UTCsetting=generative2022.050.522242.5662.469.5115.670.6386
UTC2022.050.522242.5662.469.5115.670.6386
LTMI-LGSent. Trans.=n/a2025.030.51341.361.669—0.632
GOGSent. Trans.=n/a2025.030.51341.261.8——0.626
MITVG2022.050.511441.0361.2568.4914.370.6147
V²DialSent. Trans.=ROBERTA, trained_and_evaluated_by_authors=true2025.030.50945.454.761.1—0.54
LTMIDecoder Type=Generative2019.110.507440.4461.6169.7114.930.6358
LTMIsetting=generative2022.050.507440.4461.6169.7114.930.6358
LTMI2022.050.507440.4461.6169.7114.930.6358
LTMIDecoder Type=Generative2026.050.507440.4461.6169.7114.930.6358
LTMISent. Trans.=n/a2025.030.50740.461.669.7—0.635
UTCSent. Trans.=n/a2025.030.50641.359.869.4—0.61
DAM2022.050.505140.5360.8467.9416.650.6093
DMRM2022.050.501640.1560.0267.2115.19—
KBGN2022.050.500540.460.1166.8217.540.6042
ReDANDecoder Type=Generative2019.110.500240.2759.9366.7817.40.6047
ReDANsetting=generative2022.050.500240.2759.9366.7817.40.6047
ReDAN2022.050.500240.2759.9366.7817.40.6047
ReDANDecoder Type=Generative2026.050.500240.2759.9366.7817.40.6047
CoAttDecoder Type=Generative2019.110.496440.0959.3765.9217.860.5924
CoAttsetting=generative2022.050.496440.0959.3765.9217.860.5924
CoAttre-implemented=by [19]2022.050.496440.0959.3765.9217.860.5924
CoAttDecoder Type=Generative2026.050.496440.0959.3765.9217.860.5924
HCIAEDecoder Type=Generative2019.110.490739.7258.2364.7318.430.597
HCIAEsetting=generative2022.050.490739.7258.2364.7318.430.597
HCIAEre-implemented=by [19]2022.050.490739.7258.2364.7318.430.597
HCIAEDecoder Type=Generative2026.050.490739.7258.2364.7318.430.597
Primary2022.050.490138.5459.8266.9416.6—
MNre-implemented=by [19]2022.050.479938.1857.5464.3218.60.5186
MNDecoder Type=Generative2019.110.478338.0157.4964.0818.760.5699
MNsetting=generative2022.050.478338.0157.4964.0818.760.5699
MNDecoder Type=Generative2026.050.478338.0157.4964.0818.760.5699
MST-MIXERSent. Trans.=ROBERTA, trained_and_evaluated_by_authors=true2025.030.47742.251.657.8—0.525
V²DialZero-shot=true, Sent. Trans.=OpenAI TE-32025.030.32722.141.248.1—0.32
V²DialZero-shot=true, Sent. Trans.=ROBERTa2025.030.2692030.239.3—0.333
ESPERZero-shot=true, Sent. Trans.=n/a2025.030.25714.6———0.223
ChampagneZero-shot=true, Sent. Trans.=n/a2025.030.255————0.196
VideoChat2Zero-shot=true, Sent. Trans.=ROBERTa2025.030.22312.72939.9—0.309
FROMAGEZero-shot=true, Sent. Trans.=n/a2025.030.2217.620.125.1—0.165
MST-MIXERZero-shot=true, Sent. Trans.=ROBERTa2025.030.21918.222.125.7—0.246
MOE-LLAVAZero-shot=true, Sent. Trans.=ROBERTa2025.030.20710.625.436.4—0.267
Video-ChatGPTZero-shot=true, Sent. Trans.=ROBERTa2025.030.1811022.531.5—0.248
MiniGPT-videoZero-shot=true, Sent. Trans.=ROBERTa2025.030.1467.417.426.5—0.232
Champagne/CCSent. Trans.=n/a2025.03———66.3—0.625