List your company
ResearchFormerly WizWand

Relation Extraction on TACRED v1.0 (5% train)

0.69Micro F1NLI-DeBERTa
0.271920.380460.4890.59754Sep 8, 2021Oct 20, 2021Dec 1, 2021Jan 12, 2022Feb 23, 2022Apr 6, 2022May 19, 2022
Updated 8mo ago

Evaluation results

MethodLinks
NLI-DeBERTaMethod paradigm=Indirect Supervision, Backbone architecture=DeBERTa2022.050.69——
NLI-DeBERTatraining_data_fraction=5%2021.090.690.6410.748
SURE-BART-large-cnnMethod paradigm=Summarization-based, Backbone architecture=BART-large-cnn2022.050.653——
SURE-PEGASUS-largeMethod paradigm=Summarization-based, Backbone architecture=PEGASUS-large2022.050.649——
SURE-BART-large-xsumMethod paradigm=Summarization-based, Backbone architecture=BART-large-xsum2022.050.643——
NLI-RoBERTaMethod paradigm=Indirect Supervision, Backbone architecture=RoBERTa2022.050.641——
NLI-RoBERTatraining_data_fraction=5%2021.090.6410.6040.683
SURE-BART-largeMethod paradigm=Summarization-based, Backbone architecture=BART-large2022.050.638——
IRE-RoBERTa-largeMethod paradigm=Classification-based, Backbone architecture=RoBERTa-large, Implementation=Re-implemented2022.050.636——
KnowPromptMethod paradigm=Indirect Supervision, Implementation=Re-implemented2022.050.61——
RECENTMethod paradigm=Classification-based, Implementation=Re-implemented2022.050.533——
K-AdapterMethod paradigm=Classification-based, Implementation=Re-implemented2022.050.516——
LUKEMethod paradigm=Classification-based, Implementation=Re-implemented2022.050.516——
LUKEtraining_data_fraction=5%2021.090.5160.5710.47
K-Adaptertraining_data_fraction=5%2021.090.4510.5640.376
RoBERTaMethod paradigm=Classification-based, Implementation=Re-implemented2022.050.418——
RoBERTatraining_data_fraction=5%2021.090.4180.5280.346
SpanBERTMethod paradigm=Classification-based, Implementation=Re-implemented2022.050.288——
SpanBERTtraining_data_fraction=5%2021.090.2880.3630.239