SOTAVerified

Unsupervised Pre-training

Pre-training a neural network using unsupervised (self-supervised) auxiliary tasks on unlabeled data.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 71–80 of 265 papers

TitleStatusHype
HIQL: Offline Goal-Conditioned RL with Latent States as ActionsCode1
Drop your Decoder: Pre-training with Bag-of-Word Prediction for Dense Passage RetrievalCode1
ProposalContrast: Unsupervised Pre-training for LiDAR-based 3D Object DetectionCode1
RandomRooms: Unsupervised Pre-training from Synthetic Shapes and Randomized Layouts for 3D Object DetectionCode1
SEED: Self-supervised Distillation For Visual RepresentationCode1
ELECTRIcity: An Efficient Transformer for Non-Intrusive Load MonitoringCode1
TACRED Revisited: A Thorough Evaluation of the TACRED Relation Extraction TaskCode1
Rolling-Unrolling LSTMs for Action Anticipation from First-Person VideoCode1
Accelerating Training of Transformer-Based Language Models with Progressive Layer DroppingCode0
Post Training in Deep Learning with Last KernelCode0
Show:102550
← PrevPage 8 of 27Next →
All datasetsMeaslesUCI measles

Benchmark Results

#ModelMetricClaimedVerifiedStatus
115RDLsAccuracy (%)95—Unverified
29RDLsAccuracy (%)94—Unverified
33 RMDLAccuracy (%)93—Unverified
4CNNAccuracy (%)73—Unverified
5RMDLAccuracy (%)0.1—Unverified
#ModelMetricClaimedVerifiedStatus
1RMDL (30 RDLs)Sensitivity (VEB)90.69—Unverified
2Sensitivity89.1—Unverified
3RMDL 3 RDLsSensitivity0.87—Unverified