List your company
ResearchFormerly WizWand

Language Modeling on SAMSum

31.18Perplexity(w+a)kNN-LMa (rescore)
20.50492.567164.63236.693Nov 15, 2022
Updated 8mo ago

Evaluation results

MethodLinks
(w+a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=combined, rescore=true, context_representation=LN22022.1131.18
(w+a)kNN-LMaLMa=domain-adapted GPT-2, datastore=combined, context_representation=LN22022.1131.2
(a)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=adaptation, rescore=true, context_representation=LN22022.1131.47
(a)kNN-LMaLMa=domain-adapted GPT-2, datastore=adaptation, context_representation=LN22022.1131.48
(w)kNN-LMa (rescore)LMa=domain-adapted GPT-2, datastore=pretraining, rescore=true, context_representation=LN22022.1131.92
(w)kNN-LMaLMa=domain-adapted GPT-2, datastore=pretraining, context_representation=LN22022.1131.94
LMa (only)LMa=domain-adapted GPT-22022.1132.26
(a)kNN-LMLM=GPT-2, datastore=adaptation, context_representation=original2022.1139.99
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=LN22022.1146.2
(w)kNN-LMLM=GPT-2, datastore=pretraining, context_representation=original2022.1146.27
LM (only)LM=GPT-22022.1146.86
(a)kNN (only)datastore=adaptation2022.11229.89
(w)kNN (only)datastore=pretraining2022.11298.08