Table Fact Verification on TABFACT complex (test)
85.6AccuracyPASTA
Evaluation results
| Method | Links | |
|---|---|---|
| PASTA2022.11 | 85.6 | |
| DeBERTaV32022.11 | 82.9 | |
| SaMoE2022.11 | 80.9 | |
| REASTAP2022.10 | 80.1 | |
| TAPEXpre-training_steps=50000, fine-tuning_steps=200002021.07 | 79.6 | |
| Tapex2022.11 | 79.6 | |
| TAPEX2022.10 | 79.1 | |
| DecompTaPas2022.10 | 77.4 | |
| BART2021.07 | 76 | |
| TableFormer2022.10 | 75.9 | |
| BART2022.10 | 75.7 | |
| TAPASModel=Large, Pre-training Objective=Counterfactual + Synthetic2020.10 | 75.6 | |
| Eisenschlos et al.model_category=Pre-trained Language Models2021.07 | 75.6 | |
| TaPas2022.10 | 75.6 | |
| Tapas2022.11 | 75.6 | |
| TAPASModel=Base, Pre-training Objective=Counterfactual + Synthetic2020.10 | 72.5 | |
| Schlichtkrull et al.Retrieval=Oracle2022.11 | 72.1 | |
| TAPASModel=Base, Pre-training Objective=Synthetic2020.10 | 72 | |
| TAPASModel=Base, Pre-training Objective=Counterfactual2020.10 | 68.9 | |
| TAPASModel=Base, Pre-training Objective=SQA2020.10 | 68.4 | |
| Yang et al.model_category=Pre-trained Language Models2021.07 | 67.6 | |
| ProgVGAT2022.11 | 67.6 | |
| Zhang et al.model_category=Pre-trained Language Models2021.07 | 67.2 | |
| SAT2022.10 | 67.2 | |
| SAT2022.11 | 67.2 | |
| Shi et al.model_category=Pre-trained Language Models2021.07 | 65.7 | |
| LOGICALFACTCHECKER (program from Seq2Action)2020.10 | 65.1 | |
| Zhong et al.model_category=Pre-trained Language Models2021.07 | 65.1 | |
| LFC2022.10 | 65.1 | |
| HeterTFV2022.10 | 65.1 | |
| LogicFactChecker2022.11 | 65.1 | |
| LOGICALFACTCHECKER (program from LPA)2020.10 | 64.8 | |
| TAPASModel=Base, Pre-training Objective=MASK-LM2020.10 | 63.9 | |
| LPA-Ranking w/ Discriminator (Caption)2020.10 | 58.5 | |
| LPA-Ranking2022.10 | 58.5 | |
| TABLE-BERT-Horizontal-T+F-Template2020.10 | 58.2 | |
| Chen et al.model_category=Pre-trained Language Models2021.07 | 58.2 | |
| Table-BERT2022.11 | 58.2 | |
| BERT classifier w/o Table2020.10 | 50.1 |