List your company
ResearchFormerly WizWand

Language Modeling on XSum

14.71Perplexity(w+a)kNN-LMa (rescore)
11.261634.538357.81581.0917Nov 15, 2022
Updated 8mo ago

Evaluation results

MethodLinks
(w+a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=combined, rescore=true, context_representation=LN22022.1114.71
(a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=adaptation, rescore=true, context_representation=LN22022.1114.85
(w+a)kNN-LMaLMa=domain-adapted GPT-2, datastore=combined, context_representation=LN22022.1115.2
(a)kNN-LMaLMa=domain-adapted GPT-2, datastore=adaptation, context_representation=LN22022.1115.3
(a)kNN-LMLM=GPT-2, datastore=adaptation, context_representation=original2022.1117.01
(w+a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=combined, rescore=true, context_representation=LN22022.1117.53
(a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=adaptation, rescore=true, context_representation=LN22022.1117.72
(w+a)kNN-LMaLMa=domain-adapted GPT-2, datastore=combined, context_representation=LN22022.1117.99
(a)kNN-LMaLMa=domain-adapted GPT-2, datastore=adaptation, context_representation=LN22022.1118.12
(w)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=pretraining, rescore=true, context_representation=LN22022.1118.23
(w)kNN-LMaLMa=domain-adapted GPT-2, datastore=pretraining, context_representation=LN22022.1118.42
LMa (only)LMa=domain-adapted GPT-22022.1118.95
(a)kNN-LMLM=GPT-2, datastore=adaptation, context_representation=original2022.1119.39
(w)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=pretraining, rescore=true, context_representation=LN22022.1120.98
(w)kNN-LMaLMa=domain-adapted GPT-2, datastore=pretraining, context_representation=LN22022.1121.22
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=LN22022.1121.32
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=original2022.1121.64
LMa (only)LMa=domain-adapted GPT-22022.1121.84
LM (only)LM=GPT-2, configuration=standard2022.1122.45
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=LN22022.1123.68
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=original2022.1124.03
LM (only)LM=GPT-22022.1124.87
(a)kNN (only)datastore=adaptation2022.1138.38
(a)kNN (only)datastore=adaptation2022.1147.75
(w)kNN (only)datastore=pretraining2022.1183.96
(w)kNN (only)datastore=pretraining2022.11100.92