Subjectivity Classification on Subj
99.6AccuracySimCSE-RoBERTa_base
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SimCSE-RoBERTa_baseBackbone=RoBERTa-base2022.03 | 99.6 | — | — | — | |
| SCD-BERT_baseBackbone=BERT-base2022.03 | 99.56 | — | — | — | |
| RoBERTa (Gao et al., 2021)# Params=1.0x, Evaluation Protocol=Full fine-tuning2022.12 | 97 | — | — | — | |
| ICLBackbone=Llama-2 70B, Shots=128-shots2024.05 | 96.5 | — | — | — | |
| FADS-ICLBackbone=Llama-1 30B, Shots=128-shots2024.05 | 96.4 | — | — | — | |
| FADS-ICLBackbone=Llama-2 70B, Shots=128-shots2024.05 | 96.3 | — | — | — | |
| FADS-ICLBackbone=Llama-2 13B, Shots=128-shots2024.05 | 96.2 | — | — | — | |
| TreeNet-GloVe2019.02 | 95.9 | — | — | — | |
| FADS-ICLBackbone=Llama-1 13B, Shots=128-shots2024.05 | 95.7 | — | — | — | |
| FADS-ICLBackbone=Llama-1 7B, Shots=128-shots2024.05 | 95.6 | — | — | — | |
| AdaSent2018.06 | 95.5 | — | — | — | |
| AdaSent2015.04 | 95.5 | — | — | — | |
| AdaSentNN category=Other2016.11 | 95.5 | — | — | — | |
| ADASENT2017.04 | 95.5 | — | — | — | |
| AdaSent2019.02 | 95.5 | — | — | — | |
| kNN-promptingBackbone=Llama-2 7B, Shots=128-shots2024.05 | 95.5 | — | — | — | |
| BERT [CLS]-embeddingBackbone=BERT-base, Pooling=CLS2022.03 | 95.48 | — | — | — | |
| CNN+MCFAAdditional Context=Translation, N=ensemble2018.06 | 95.2 | — | — | — | |
| FADS-ICLBackbone=Llama-2 7B, Shots=128-shots2024.05 | 95.2 | — | — | — | |
| TopCNNAdditional Context=Topic, Granularity=ensemble2018.06 | 95 | — | — | — | |
| VLAWE2019.02 | 95 | — | — | — | |
| IS-BERT_baseBackbone=BERT-base2022.03 | 94.96 | — | — | — | |
| CNN+B1Additional Context=Translation, N=ensemble2018.06 | 94.9 | — | — | — | |
| MC-QTd*=48002018.05 | 94.8 | — | — | — | |
| CNN+B2Additional Context=Translation, N=ensemble2018.06 | 94.8 | — | — | — | |
| CNN+MCFAAdditional Context=Translation, N=102018.06 | 94.8 | — | — | — | |
| byte mLSTMd*=40962018.05 | 94.7 | — | — | — | |
| CNN+MCFAAdditional Context=Translation, N=12018.06 | 94.7 | — | — | — | |
| CNN+B1Additional Context=Translation, N=12018.06 | 94.6 | — | — | — | |
| CNN+B2Additional Context=Translation, N=12018.06 | 94.6 | — | — | — | |
| BYTE MLSTMClassifier=Logistic Regression, Penalty=L12017.04 | 94.6 | — | — | — | |
| kNN-promptingBackbone=Llama-2 13B, Shots=128-shots2024.05 | 94.6 | — | — | — | |
| SBERT2025.12 | 94.5 | — | — | — | |
| Avg. BERT embeddingsBackbone=BERT-base, Pooling=Avg2022.03 | 94.37 | — | — | — | |
| skip-thoughtsd*=48002018.05 | 94.2 | — | — | — | |
| BRNN2015.04 | 94.2 | — | — | — | |
| RoBERTa [CLS]-embeddingBackbone=RoBERTa-base, Pooling=CLS2022.03 | 94.15 | — | — | — | |
| Shared-Layer ArchitectureMulti-Task=true, Fine-Tuning=true2016.05 | 94.1 | — | — | — | |
| Multi-TaskNN category=RNN2016.11 | 94.1 | — | — | — | |
| DARLM2019.02 | 94.1 | — | — | — | |
| CNN+B2Additional Context=Translation, N=102018.06 | 94 | — | — | — | |
| BLSTM-2DCNNNN category=ours2016.11 | 94 | — | — | — | |
| BLSTM-2DCNN2019.02 | 94 | — | — | — | |
| SimCSE-BERT_baseBackbone=BERT-base2022.03 | 93.96 | — | — | — | |
| MVCNNNN category=CNN2016.11 | 93.9 | — | — | — | |
| USE_T+DAN (w2v w.e.)Universal Encoder=Transformer (USE_T), Transfer Level=Sentence & Word, Embedding Initialization=word2vec skip-gram, Transfer Task Model=DAN2018.03 | 93.9 | — | — | — | |
| USE TUniversal Encoder=Transformer (USE_T), Transfer Level=Sentence, Embedding Initialization=None, Transfer Task Model=None2018.03 | 93.87 | — | — | — | |
| à la carten=3, d*=48002018.05 | 93.8 | — | — | — | |
| CNN+B1Additional Context=Translation, N=102018.06 | 93.8 | — | — | — | |
| Capsule-B2018.03 | 93.8 | — | — | — | |
| RNN2015.04 | 93.7 | — | — | — | |
| BLSTM-2DPoolingNN category=ours2016.11 | 93.7 | — | — | — | |
| SKIPTHOUGHTLayer Normalization=True2017.04 | 93.7 | — | — | — | |
| DC-TreeLSTM2019.02 | 93.7 | — | — | — | |
| SCD-RoBERTa_baseBackbone=RoBERTa-base2022.03 | 93.67 | — | — | — | |
| CNN-AnaNN category=CNN2016.11 | 93.66 | — | — | — | |
| USE_T+DAN (lrn w.e.)Universal Encoder=Transformer (USE_T), Transfer Level=Sentence, Embedding Initialization=learned on transfer task, Transfer Task Model=DAN2018.03 | 93.66 | — | — | — | |
| MNB2014.08 | 93.6 | — | — | — | |
| F-Dropout2014.08 | 93.6 | — | — | — | |
| CNN-LSTMd*=48002018.05 | 93.6 | — | — | — | |
| MNB2015.04 | 93.6 | — | — | — | |
| SkipThoughtData Source Type=Ordered Sentences2016.02 | 93.6 | — | — | — | |
| combine-skipNN category=Other2016.11 | 93.6 | — | — | — | |
| Combine-skip2019.02 | 93.6 | — | — | — | |
| Combine-skip + NB2019.02 | 93.6 | — | — | — | |
| Skip-thoughtModel=Skip-thought2022.03 | 93.6 | — | — | — | |
| kNN-promptingBackbone=Llama-1 30B, Shots=128-shots2024.05 | 93.6 | — | — | — | |
| USE_T+CNN (w2v w.e.)Universal Encoder=Transformer (USE_T), Transfer Level=Sentence & Word, Embedding Initialization=word2vec skip-gram, Transfer Task Model=CNN2018.03 | 93.58 | — | — | — | |
| à la carten=2, d*=32002018.05 | 93.5 | — | — | — | |
| BLSTM-AttNN category=ours2016.11 | 93.5 | — | — | — | |
| BLSTM-Att2019.02 | 93.5 | — | — | — | |
| CNN-non-staticword vectors=fine-tuned pre-trained (word2vec)2014.08 | 93.4 | — | — | — | |
| G-Dropout2014.08 | 93.4 | — | — | — | |
| CNN2018.06 | 93.4 | — | — | — | |
| TopCNNAdditional Context=Topic, Granularity=word2018.06 | 93.4 | — | — | — | |
| TopCNNAdditional Context=Topic, Granularity=sentence2018.06 | 93.4 | — | — | — | |
| CNN2015.04 | 93.4 | — | — | — | |
| CNN-non-staticNN category=CNN2016.11 | 93.4 | — | — | — | |
| CNN2017.04 | 93.4 | — | — | — | |
| CNNmode=non-static2018.03 | 93.4 | — | — | — | |
| CNN2019.02 | 93.4 | — | — | — | |
| kNN-promptingBackbone=Llama-1 7B, Shots=128-shots2024.05 | 93.4 | — | — | — | |
| USE_T+CNN (lrn w.e.)Universal Encoder=Transformer (USE_T), Transfer Level=Sentence, Embedding Initialization=learned on transfer task, Transfer Task Model=CNN2018.03 | 93.36 | — | — | — | |
| Capsule-A2018.03 | 93.3 | — | — | — | |
| COV + BOW2019.02 | 93.3 | — | — | — | |
| USE_D+CNN (w2v w.e.)Universal Encoder=Deep Averaging Network (USE_D), Transfer Level=Sentence & Word, Embedding Initialization=word2vec skip-gram, Transfer Task Model=CNN2018.03 | 93.24 | — | — | — | |
| CNN-multichannelword vectors=pre-trained (word2vec), channels=22014.08 | 93.2 | — | — | — | |
| NBSVM2014.08 | 93.2 | — | — | — | |
| DSCNN2018.06 | 93.2 | — | — | — | |
| NB-SVM2015.04 | 93.2 | — | — | — | |
| CNN-MCNN category=CNN2016.11 | 93.2 | — | — | — | |
| DSCNNNN category=Other2016.11 | 93.2 | — | — | — | |
| NBSVM2017.04 | 93.2 | — | — | — | |
| COV + Mean + BOW2019.02 | 93.2 | — | — | — | |
| USE_D+DAN (w2v w.e.)Universal Encoder=Deep Averaging Network (USE_D), Transfer Level=Sentence & Word, Embedding Initialization=word2vec skip-gram, Transfer Task Model=DAN2018.03 | 93.12 | — | — | — | |
| COV + Mean2019.02 | 93.1 | — | — | — | |
| SCLie2025.12 | 93.1 | — | — | — | |
| CNN-staticword vectors=static pre-trained (word2vec)2014.08 | 93 | — | — | — | |
| DCNN2014.08 | 93 | — | — | — | |
| CNNmode=static2018.03 | 93 | — | — | — |