Our new X account is live! Follow @wizwand_team for updates

Sentiment Classification on MR

90.8Accuracy

RoBERTa (Gao et al., 2021)

Updated 2d ago

Evaluation Results

Method	Links
RoBERTa (Gao et al., 2021) 2022.12		90.8
BERT+SCM 2023.08		88.16
BERT (2019) 2023.08		87.65
SBERT 2025.12		84.9
Kim et al., 2019 2023.08		83.8
NPM 2022.12		83.7
PICL 2023.05		83.6
BILSTM+SCM 2023.08		83.41
BILSTM (1997) 2023.08		83.27
CNN+MCFA 2018.06		83.2
AdaSent 2018.06		83.1
AdaSent 2015.04		83.1
AdaSent 2016.11		83.1
TopCNN 2018.06		83
CNN+MCFA 2018.06		82.7
CNN+B1 2018.06		82.6
CNN+MCFA 2018.06		82.3
BRNN 2015.04		82.3
BLSTM-2DCNN 2016.11		82.3
DSCNN 2018.06		82.2
CNN+B2 2018.06		82.2
SCD-RoBERTa_base 2022.03		82.17
CNN+B2 2018.06		82.1
CNN+B2 2018.06		82.1
Dep-CNN 2018.06		81.9
CNN+B1 2018.06		81.9
Liu et al., 2017a 2023.08		81.9
BERT [CLS]-embedding 2022.03		81.83
TopCNN 2018.06		81.7
CNN-Rule-q 2016.03		81.7
Liu et al., 2017b 2023.08		81.7
RoBERTa 2022.12		81.7
CNN-Rule-p 2016.03		81.6
USE_T+CNN (lrn w.e.) 2018.03		81.59
CNN-non-static 2014.08		81.5
CNN 2018.06		81.5
CNN 2015.04		81.5
DSCNN 2016.11		81.5
BLSTM-2DPooling 2016.11		81.5
USE T 2018.03		81.44
CNN+B1 2018.06		81.4
USE_T+DAN (lrn w.e.) 2018.03		81.36
USE_T+DAN (w2v w.e.) 2018.03		81.32
TopCNN 2018.06		81.3
CNN 2016.03		81.3
RoBERTa [CLS]-embedding 2022.03		81.27
USE_T+CNN (w2v w.e.) 2018.03		81.18
CNN-multichannel 2014.08		81.1
CNN-multichannel 2016.03		81.1
IS-BERT_base 2022.03		81.09
CNN-Ana 2016.11		81.02
CNN-static 2014.08		81
BLSTM-Att 2016.11		81
RoBERTa (Gao et al., 2021) 2022.12		80.8
SimCSE-BERT_base 2022.03		80.74
LM+TSED+PT+2L 2019.05		80.33
STM+TSED+PT+2L 2019.05		80.09
BLSTM 2016.11		80
SCLie 2025.12		79.8
SCNN 2025.12		79.7
Sent-Parser 2014.08		79.5
NBSVM 2014.08		79.4
NB-SVM 2015.04		79.4
NLM+TSED+PT+2L 2019.05		79.33
LSTM-400 2019.05		79.28
F-Dropout 2014.08		79.1
CNN-400 2019.05		79.07
LM+TSED+PT+2L 2019.05		79.06
MV-RNN 2014.08		79
MNB 2014.08		79
G-Dropout 2014.08		79
MNB 2015.04		79
MV-RecNN 2015.04		79
G-Dropout 2016.03		79
MVRNN (2012) 2023.08		79
NLM+TSED 2019.05		78.89
STM+TSED 2019.05		78.85
STM+TSED+PT+2L 2019.05		78.77
TreeLSTM (2015) 2023.08		78.7
Avg. BERT embeddings 2022.03		78.66
LM+TSED 2019.05		78.61
NLM+TSED+PT+2L 2019.05		78.6
USE_D+CNN (lrn w.e.) 2018.03		78.49
USE_D+CNN (w2v w.e.) 2018.03		78.2
GPT-2 kNN-LM 2022.12		78.2
NLM+TSED 2019.05		78.1
CNN-50 2019.05		78.07
STM+TSED 2019.05		77.81
CCAE 2014.08		77.8
RAE 2014.08		77.7
RAE 2015.04		77.7
LM+TSED 2019.05		77.63
USE_D+DAN (lrn w.e.) 2018.03		77.57
DPCLie 2025.12		77.4
Tree-CRF 2014.08		77.3
GloVe embeddings (avg.) 2022.03		77.25
CBOW 2015.04		77.2
RNN 2015.04		77.2
DPCNN 2025.12		77.2
LSTM-50 2019.05		77.16

Showing 100 of 148 rows