List your company
ResearchFormerly WizWand

Language Modeling on BookSum

19.35Perplexity(w)kNN-LMa (rescore)
15.729240.169664.6189.0504Nov 15, 2022
Updated 8mo ago

Evaluation results

MethodLinks
(w)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=pretraining, rescore=true, context_representation=LN22022.1119.35
(w+a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=combined, rescore=true, context_representation=LN22022.1119.42
(w)kNN-LMaLMa=domain-adapted GPT-2, datastore=pretraining, context_representation=LN22022.1119.72
(w+a)kNN-LMaLMa=domain-adapted GPT-2, datastore=combined, context_representation=LN22022.1119.72
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=LN22022.1120.09
(a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=adaptation, rescore=true, context_representation=LN22022.1120.09
(a)kNN-LMaLMa=domain-adapted GPT-2, datastore=adaptation, context_representation=LN22022.1120.22
LMa (only)LMa=domain-adapted GPT-22022.1120.45
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=original2022.1120.5
(a)kNN-LMLM=GPT-2, datastore=adaptation, context_representation=original2022.1120.6
LM (only)LM=GPT-22022.1121.15
(w)kNN (only)datastore=pretraining2022.1164.92
(a)kNN (only)datastore=adaptation2022.11109.87