List your company
ResearchFormerly WizWand

Language Modeling on SciQ

11.95Perplexity(w+a)kNN-LMa (rescore)
10.115222.500134.88547.2699Nov 15, 2022
Updated 8mo ago

Evaluation results

MethodLinks
(w+a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=combined, rescore=true, context_representation=LN22022.1111.95
(w+a)kNN-LMaLMa=domain-adapted GPT-2, datastore=combined, context_representation=LN22022.1112.15
(a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=adaptation, rescore=true, context_representation=LN22022.1112.72
(a)kNN-LMaLMa=domain-adapted GPT-2, datastore=adaptation, context_representation=LN22022.1112.88
(w)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=pretraining, rescore=true, context_representation=LN22022.1114.22
(w)kNN-LMaLMa=domain-adapted GPT-2, datastore=pretraining, context_representation=LN22022.1114.62
(a)kNN-LMLM=GPT-2, datastore=adaptation, context_representation=original2022.1114.71
LMa (only)LMa=domain-adapted GPT-22022.1116.09
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=LN22022.1118.5
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=original2022.1119.19
LM (only)LM=GPT-22022.1122.15
(w)kNN (only)datastore=pretraining2022.1146.86
(a)kNN (only)datastore=adaptation2022.1157.82