List your company
ResearchFormerly WizWand

Text Classification on SST-5 (test)

56.21AccuracyPRL
35.6741.002546.33551.6675May 18, 2018Sep 7, 2019Dec 28, 2020Apr 19, 2022Aug 10, 2023Nov 29, 2024Mar 22, 2026
Updated 7mo ago

Evaluation results

MethodLinks
PRL2025.1256.21
BCN with SuBiLSTM-Tied+CoVeArchitecture=BCN, Encoder=SuBiLSTM-Tied, Embeddings=CoVe2018.0556.2
PLR-4Model=Llama3.1-8B-Instruct, k=322026.0355.96
DE2025.1255.25
T-few 3Bsamples per class=82022.1055
StaticModel=Qwen2.5-7B-Instruct, k=42026.0354.78
BCN with SuBiLSTM+CoVeArchitecture=BCN, Encoder=SuBiLSTM, Embeddings=CoVe2018.0554.5
BCN with 2-layer BiLSTM+CoVeArchitecture=BCN, Encoder=2-layer BiLSTM, Embeddings=CoVe2018.0554
APO2025.1253.94
GA2025.1253.76
BCN+Char+CoVeArchitecture=BCN, Embeddings=Char, CoVe2018.0553.7
BCN with BiLSTM+CoVeArchitecture=BCN, Encoder=BiLSTM, Embeddings=CoVe2018.0553.6
BCN with 2-layer BiLSTMArchitecture=BCN, Encoder=2-layer BiLSTM2018.0553.5
BCN with SuBiLSTM-TiedArchitecture=BCN, Encoder=SuBiLSTM-Tied2018.0553.4
PIAST (E)variant=Ensemble2025.1253.33
PIAST2025.1253.27
BCN with BiLSTMArchitecture=BCN, Encoder=BiLSTM2018.0553.2
BCN with SuBiLSTMArchitecture=BCN, Encoder=SuBiLSTM2018.0553.2
PIAST (LOO)variant=Leave-one-out2025.1253.18
NTIArchitecture=NTI2018.0553.1
TE-LSTMArchitecture=TE-LSTM2018.0552.6
MI2025.1252.31
NI2025.1251.9
Fine-tuned BERTSNN=false, Spike=false, PE=false, Param (M)=109.82024.0550.41
PIAST (I)variant=Iterative2025.1249.79
APE2025.1249.37
Ensemble of SetFit + OCDsamples per class=82022.1048.6
Mean of 5 OCD instancessamples per class=82022.1047.9
SetFit + OCDsamples per class=82022.1047.8
Our method (Weight Net)Imbalance Factor=500:752022.0844.95
Our method (Weight Net)Imbalance Factor=500:602022.0844.79
Our method (Weight Net)Imbalance Factor=500:502022.0844.68
Constraint-based re-weightingImbalance Factor=500:502022.0844.62
Fine-Tuned BERT+CEImbalance Factor=500:602022.0844.06
Fine-Tuned BERTImbalance Factor=500:752022.0843.76
Fine-Tuned BERT+CEImbalance Factor=500:752022.0843.71
Spikformer w/ CPG-PESNN=true, Spike=true, PE=true, Param (M)=110.42024.0543.62
SetFitsamples per class=82022.1043.6
Logit AdjustmentImbalance Factor=500:752022.0843.52
Fine-Tuned BERTImbalance Factor=500:602022.0843.25
LDAM-DRWImbalance Factor=500:752022.0843.2
ProportionImbalance Factor=500:752022.0842.78
Probabilistic Transformer2023.1142.77
Fine-Tuned BERTImbalance Factor=500:502022.0842.7
Spikformer w/ Float-PESNN=true, Spike=false, PE=true, Param (M)=109.82024.0542.58
ProportionImbalance Factor=500:602022.0842.36
Spikformer w/ Random-PESNN=true, Spike=true, PE=true, Param (M)=110.42024.0541.86
Spikformer w/o PESNN=true, Spike=false, PE=false, Param (M)=109.82024.0541.84
ProportionImbalance Factor=500:502022.0841.6
LDAM-DRWImbalance Factor=500:602022.0840.9
LDAM-DRWImbalance Factor=500:502022.0840.81
Hu et al.'s+RegularizationImbalance Factor=500:502022.0840.14
Transformer2023.1140.13
Hu et al.'sImbalance Factor=500:502022.0839.82
Logit AdjustmentImbalance Factor=500:602022.0839.52
Vanilla BERTImbalance Factor=500:752022.0836.99
Vanilla BERTImbalance Factor=500:602022.0836.75
Logit AdjustmentImbalance Factor=500:502022.0836.55
Vanilla BERTImbalance Factor=500:502022.0836.46
Fine-Tuned BERT+CEImbalance Factor=500:502022.0836.46