List your company

Math Reasoning on ASDiv A (test)

91AccuracyATHENA
50.12860.73971.3581.961Oct 28, 2022Dec 28, 2022Feb 28, 2023May 1, 2023Jul 1, 2023Sep 1, 2023Nov 2, 2023

Evaluation results

MethodLinks
ATHENABackbone=RoBERTa-large2023.1191
CoReBackbone=GPT-J 12B, Evaluation Protocol=Zero-shot2022.1090.5
DeductReasonerBackbone=RoBERTa-large2023.1189.1
ATHENABackbone=RoBERTa-base2023.1186.4
DeductReasonerBackbone=RoBERTa-base2023.1183.1
R-Graph-to-TreeBackbone=RoBERTa-base2023.1182.2
R-GTSBackbone=RoBERTa-base2023.1181.2
Graph-to-TreeBackbone=Random embedding2023.1177.4
Previous SoTAEvaluation Protocol=Fine-tune2022.1075.3
R-TransformerBackbone=RoBERTa-base2023.1172.1
GTSBackbone=Random embedding2023.1171.4
Self-ConsistencyBackbone=GPT-J 6B, Evaluation Protocol=Zero-shot2022.1063.7
TransformerBackbone=Random embedding2023.1159.3
Generator onlyBackbone=GPT-J 6B, Evaluation Protocol=Zero-shot2022.1051.7