List your company
ResearchFormerly WizWand

Text Classification on R52 (test)

96.6AccuracyBertGCN
87.229689.662392.09594.5277Sep 26, 2019Oct 10, 2020Oct 25, 2021Nov 9, 2022Nov 24, 2023Dec 8, 2024Dec 23, 2025
Updated 8mo ago

Evaluation results

MethodLinks
BertGCNBackbone=BERT-base2021.0596.6
BertGATBackbone=BERT-base2021.0596.5
BERTModel variant=BERT-base2021.0596.4
TextGSL2025.1296.26
RoBERTa2021.0596.2
ROBERTAGCNBackbone=RoBERTa2021.0596.1
ROBERTAGATBackbone=RoBERTa2021.0596.1
LDGCN2025.1295.71
TextSSL2025.1295.48
TextING2025.1295.41
TensorGCN2025.1295.05
HyperGAT2025.1294.98
UGformerModel Variant=Variant 2, GNN layer usage=with GNN2019.0994.71
S2GC2021.0594.5
CGA2TC2025.1294.47
GTG2025.1294.46
TextQGNN2019.0994.45
SGC2021.0594
DHTG2025.1293.93
UGformer (w/o GNN)Model Variant=Variant 2, GNN layer usage=without GNN2019.0993.85
TextGCN2021.0593.6
TextGCN2019.0993.56
TextGCN2025.1293.56
TextING2019.0993.34
fastText2019.0992.81
Drop Clause Tsetlin Machinemode=DC (Drop Clause)2021.0592.75
BiLSTM2021.0590.54
Bi-LSTM2019.0990.54
BERT2021.0589.66
CNN2021.0587.59