List your company

Factual Consistency Evaluation on SummaC

88.4CGSALIGN
43.1654.90566.6578.395May 26, 2023Jun 1, 2023Jun 8, 2023Jun 15, 2023Jun 22, 2023Jun 29, 2023Jul 6, 2023

Evaluation results

MethodLinks
ALIGNModel Scale=large2023.0788.474.692.594.992.391.389
ALIGNSCORE-largeType=Ours2023.0586.475.892.493.791.791.488.6
UniEvalType=Misc2023.0584.765.593.489.986.38884.6
UniEval2023.0784.765.593.489.986.38884.6
ALIGNSCORE-baseType=Ours2023.0583.779.487.893.389.990.587.4
QAFactEvalType=QA2023.0583.466.186.489.288.189.483.8
QAFactEval2023.0783.466.186.489.288.189.483.8
ALIGNModel Scale=base2023.0780.676.187.593.188.689.585.9
ALIGNSCORE-baseType=Ours2023.0577.872.278.987.483.783.680.6
UniEvalType=Misc2023.0577.161.285.384.779.480.978.1
CTCType=Misc2023.0576.565.989.582.685.687.381.2
CTC2023.0776.565.989.582.685.687.381.2
ALIGNSCORE-largeType=Ours2023.0575708889.283.486.382
BARTScoreType=Misc2023.0574.362.691.782.385.988.580.9
BARTScore2023.0774.362.691.782.385.988.580.9
SummaC-ZSType=NLI2023.0573.65887.583.785.885.379
BLEUType=Similarity Matching2023.0571.855.886.97583.884.576.3
ROUGE-2Type=Similarity Matching2023.0570.565.983.77687.285.378.1
ROUGE-LType=Similarity Matching2023.0570.262.981.976.387.385.377.3
ROUGE-1Type=Similarity Matching2023.0569.764.582.575.887.28577.4
CTCType=Misc2023.0569.161.782.177.678.480.574.9
SummaC-CONVType=NLI2023.0567.270.381.892.386.188.581
FactCCType=Misc2023.0564.955.178.572.771.869.868.8
FactCCType=Misc2023.0564.955.178.572.771.869.868.8
BLEUType=Similarity Matching2023.0564.255.278.36777.679.370.3
BERTScoreType=Similarity Matching, Backbone=DeBERTa-XLarge-MNLI2023.0563.14985.370.979.684.972.1
BERTScore2023.0763.14985.370.979.684.972.1
SummaC-ZSType=NLI2023.0562.657.88182.877.878.173.4
ROUGE-LType=Similarity Matching2023.0561.557.47467.779.778.869.8
ROUGE-2Type=Similarity Matching2023.0561.262.275.167.878.878.870.7
ROUGE-1Type=Similarity Matching2023.0561.162.474.4688079.170.8
BLEURTType=Regression, Model=BLEURT-202023.0560.864.776.759.771.182.569.2
BLEURT2023.0760.864.776.759.771.182.569.2
QuestEvalType=QA2023.0560.463.67774.274.385.872.5
SummaC-CONVType=NLI2023.0559.866.473.789.279.88175
BLEURTType=Regression2023.0557.758.76956.263.774.963.4
BARTScoreType=Misc2023.0556.958.784.673.379.678.371.9
SimCSEType=Similarity Matching, Model=sup-simcse-roberta-large2023.0556.262.275.25977.274.867.4
SimCSEType=Similarity Matching2023.0554.457.368.957.371.368.562.9
BLANCType=Misc2023.0554.153.574.756.468.683.465.1
FEQAType=QA2023.0553.747.654.347.948.837.248.3
QuestEvalType=QA2023.0553.157.669.366.869.877.765.7
BERTScoreType=Similarity Matching2023.0552.74976.965.372.778.565.8
DAEType=NLI2023.0552.476.772.854.266.178.966.8
DAEType=NLI2023.0552.476.772.854.266.178.966.8
FEQAType=QA2023.0551.949.553.746.651.441.449.1
NER-OverlapType=Similarity Matching2023.0551.164.972.149.856.668.160.4
NER-OverlapType=Similarity Matching2023.0551.164.972.149.856.668.160.4
QAFactEvalType=QA2023.0550.661.260.273.854.974.962.6
BLANCType=Misc2023.0549.85266.355.758.378.460.1
MNLIType=NLI2023.054648.746.352.250.755.249.8
MNLIType=NLI, Model=RoBERTa-large2023.0544.946.64548.343.559.347.9