List your company
ResearchFormerly WizWand

Language Modeling on MSC Session 5 1.0 (val)

8.99PerplexitySumMem-MSC 2.7B
8.9329.32359.71510.1065Jul 15, 2021
Updated 8mo ago

Evaluation results

MethodLinks
SumMem-MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Retrieval Strategy=FiD2021.078.99
SumMem-MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Retrieval Strategy=FiD-RAG2021.078.99
SumMem-MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Context Truncation=10242021.079
MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Retrieval Strategy=FiD2021.079.04
MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Retrieval Strategy=FiD-RAG2021.079.04
MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Context Truncation=10242021.079.08
SumMem-MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Retrieval Strategy=RAG2021.079.09
MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Retrieval Strategy=RAG2021.079.1
MSC 2.7BParameters=2.7B, Architecture=encoder-decoder, Context Truncation=1282021.079.23
BST 2.7BParameters=2.7B, Architecture=encoder-decoder2021.0710.44