List your company
ResearchFormerly WizWand

Natural Language Inference on ANLI R3 (test)

44.7AccuracyGLaM
15.68423.21730.7538.283Dec 13, 2021Jun 14, 2022Dec 15, 2022Jun 16, 2023Dec 17, 2023Jun 17, 2024Dec 18, 2024
Updated 8mo ago

Evaluation results

MethodLinks
GLaMModel Size=64B/64E, Evaluation Protocol=Few-shot, Shots=42021.1244.7
GLaMModel Size=64B/64E, Evaluation Protocol=Zero-shot2021.1241.3
GLaMModel Size=64B/64E, Evaluation Protocol=One-shot2021.1240.8
GPT-3Model Size=175B, Evaluation Protocol=Few-shot, Shots=502021.1240.2
GPT-3Model Size=175B, Evaluation Protocol=One-shot2021.1235.1
GPT-3Model Size=175B, Evaluation Protocol=Zero-shot2021.1234.5
ChaosNLI HJDEvaluation Model=RoBERTa2024.1233.8
ChaosNLI HJDEvaluation Model=BERT2024.1233.2
VariErr distributionEvaluation Model=RoBERTa2024.1232.1
Mixtral (MJD) + human explanationsEvaluation Model=BERT, Explanations=Human2024.1232
VariErr distributionEvaluation Model=BERT2024.1231.9
VariErr Label-GuidedEvaluation Model=BERT, Explanations=Model-generated2024.1231.7
MNLI Label-GuidedEvaluation Model=BERT, Explanations=Model-generated2024.1230.3
Mixtral (MJD) + human explanationsEvaluation Model=RoBERTa, Explanations=Human2024.1230
VariErr Label-GuidedEvaluation Model=RoBERTa, Explanations=Model-generated2024.1229.6
MNLI-FT-LMEvaluation Model=BERT2024.1229.3
MNLI Label-GuidedEvaluation Model=RoBERTa, Explanations=Model-generated2024.1229.2
MNLI distributionEvaluation Model=RoBERTa2024.1228.1
MNLI distributionEvaluation Model=BERT2024.1227.9
MNLI-FT-LMEvaluation Model=RoBERTa2024.1225.7
Label-FreeEvaluation Model=BERT, Explanations=Model-generated2024.1225.5
Mixtral (MJD)Evaluation Model=BERT, Explanations=None2024.1224.6
Mixtral (MJD)Evaluation Model=RoBERTa, Explanations=None2024.1224.3
Label-FreeEvaluation Model=RoBERTa, Explanations=Model-generated2024.1224.3
Out-of-the-box LMEvaluation Model=BERT2024.1219.7
Out-of-the-box LMEvaluation Model=RoBERTa2024.1216.8