ALBERT

E435869

ALBERT is a lightweight, parameter-efficient variant of the BERT language model designed to achieve strong natural language understanding performance with reduced memory and computation costs.

All labels observed (1)

Label Occurrences
ALBERT canonical 1

How this entity was disambiguated

Statements (48)

Predicate Object
instanceOf BERT variant
language model
neural network model
transformer-based model
achieves state-of-the-art results on several benchmarks at release
acronymFor A Lite BERT NERFINISHED
aimsTo maintain strong performance
reduce computation cost
reduce memory usage
basedOn BERT NERFINISHED
compatibleWith Hugging Face Transformers NERFINISHED
describedInPaper ALBERT: A Lite BERT for Self-supervised Learning of Language Representations NERFINISHED
designedFor natural language understanding
evaluatedOn GLUE benchmark NERFINISHED
RACE dataset NERFINISHED
SQuAD NERFINISHED
fullName A Lite BERT NERFINISHED
hasArchitecture Transformer NERFINISHED
hasFeature cross-layer parameter sharing
smaller embedding size with projection
hasObjective sentence-order prediction
hasOpenSourceImplementation Yes
hasProperty computationally-efficient
lightweight
memory-efficient
parameter-efficient
hasVariant ALBERT-base NERFINISHED
ALBERT-large NERFINISHED
ALBERT-xlarge NERFINISHED
ALBERT-xxlarge NERFINISHED
implementedIn TensorFlow NERFINISHED
introducedBy Google Research NERFINISHED
Toyota Technological Institute at Chicago NERFINISHED
introducedIn 2019
language English (pretrained models)
paperAuthorsInclude Kevin Gimpel NERFINISHED
Mingda Chen NERFINISHED
Piyush Sharma NERFINISHED
Radu Soricut NERFINISHED
Sebastian Goodman NERFINISHED
Zhenzhong Lan NERFINISHED
replacesObjective next sentence prediction
supports sentence-level tasks
token-level tasks
trainedWith masked language modeling
uses self-attention
usesTechnique factorized embedding parameterization
parameter sharing across layers

How these facts were elicited

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.