gptkbp:instanceOf
|
large language model
|
gptkbp:architecture
|
gptkb:transformation
|
gptkbp:author
|
gptkb:Furu_Wei
gptkb:Hangbo_Bao
gptkb:Li_Dong
gptkb:Ming_Zhou
gptkb:Nan_Yang
gptkb:Wenhui_Wang
|
gptkbp:availableOn
|
gptkb:Hugging_Face
|
gptkbp:basedOn
|
Transformer architecture
|
gptkbp:designedFor
|
natural language processing
|
gptkbp:developedBy
|
gptkb:Microsoft_Research
|
https://www.w3.org/2000/01/rdf-schema#label
|
MiniLM
|
gptkbp:language
|
English
|
gptkbp:notableFor
|
high performance
small model size
|
gptkbp:notablePublication
|
gptkb:MiniLM:_Deep_Self-Attention_Distillation_for_Task-Agnostic_Compression_of_Pre-Trained_Transformers
|
gptkbp:openSource
|
true
|
gptkbp:parameter
|
33M
|
gptkbp:predecessor
|
gptkb:BERT
|
gptkbp:publishedIn
|
gptkb:NeurIPS_2020
|
gptkbp:relatedTo
|
gptkb:DistilBERT
gptkb:TinyBERT
|
gptkbp:releaseYear
|
2020
|
gptkbp:trainer
|
knowledge distillation
|
gptkbp:usedFor
|
question answering
semantic search
text classification
sentence embeddings
|
gptkbp:variant
|
gptkb:MiniLM-L12-H384-uncased
gptkb:MiniLM-L12-H768-uncased
gptkb:MiniLM-L6-H384-uncased
|
gptkbp:bfsParent
|
gptkb:transformation
|
gptkbp:bfsLayer
|
5
|