Text Classification on R52
96.65AccuracyG-Loss
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| G-LossBackbone=RoBERTa-large2026.04 | 96.65 | — | — | |
| Bert-GCN2026.04 | 96.6 | — | — | |
| BERT w/ shuf. augm.Setting=Inductive2021.09 | 96.48 | 0.18 | — | |
| G-LossBackbone=BERT-base2026.04 | 96.48 | — | — | |
| G-Loss-O + CEBackbone=RoBERTa-large2026.04 | 96.44 | — | 84.34 | |
| BERT-base2026.04 | 96.4 | — | — | |
| GLoss-O + CEBackbone=BERT-base-uncased2026.04 | 96.37 | — | 84.55 | |
| CEBackbone=BERT-base-uncased2026.04 | 96.29 | — | 84.42 | |
| RoBERTa-large2026.04 | 96.2 | — | — | |
| SCL + CEBackbone=BERT-base-uncased2026.04 | 96.18 | — | 83.75 | |
| BERTSetting=Inductive2021.09 | 96.17 | 0.33 | — | |
| SCL + CEBackbone=DistilBERT-base2026.04 | 96.17 | — | 83.82 | |
| G-Loss-SQRT + CEBackbone=RoBERTa-large2026.04 | 96.14 | — | 84.31 | |
| GLoss-SQRT + CEBackbone=DistilBERT-base2026.04 | 96.13 | — | 83.93 | |
| RoBerta-GCN2026.04 | 96.1 | — | — | |
| GLoss-O + CEBackbone=DistilBERT-base2026.04 | 96.07 | — | 84.44 | |
| GLoss-SQRT + CEBackbone=BERT-base-uncased2026.04 | 96.06 | — | 83.78 | |
| BERTPre-training=true, Training=true2026.06 | 96 | — | — | |
| SCL + CEBackbone=RoBERTa-large2026.04 | 95.95 | — | 82.33 | |
| CEBackbone=RoBERTa-large2026.04 | 95.91 | — | 83.71 | |
| CEBackbone=DistilBERT-base2026.04 | 95.78 | — | 83.64 | |
| TMPre-training=NTM with BERT-base2026.06 | 95.44 | — | — | |
| DistilBERTSetting=Inductive2021.09 | 95.34 | 0.08 | — | |
| TMPre-training=NTM with BERT-large2026.06 | 95.27 | — | — | |
| DADGNNSetting=Inductive2021.09 | 95.16 | 0.22 | — | |
| TensorGCNSetting=Transductive2021.09 | 95.05 | — | — | |
| TensorGCN2026.04 | 95.05 | — | — | |
| HyperGATSetting=Inductive2021.09 | 94.98 | 0.27 | — | |
| fastTexth=102026.04 | 94.74 | — | — | |
| BERT w/o pos embSetting=Inductive2021.09 | 94.7 | 0.27 | — | |
| S2GC2021.04 | 94.5 | — | — | |
| LLM-Guided TM2026.04 | 94.45 | — | — | |
| HeteGCNSetting=Transductive2021.09 | 94.35 | 0.25 | — | |
| Lguided-BERT-3layers=32021.04 | 94.32 | — | — | |
| BERT-large2026.06 | 94.32 | — | — | |
| Lguided-BERT-1layers=12021.04 | 94.26 | — | — | |
| BERT2026.04 | 94.26 | — | — | |
| BERT-base2026.06 | 94.26 | — | — | |
| SGCSetting=Transductive2021.09 | 94 | 0.2 | — | |
| WideMLP-2Setting=Inductive2021.09 | 93.98 | 0.23 | — | |
| WideMLPSetting=Inductive2021.09 | 93.89 | 0.16 | — | |
| HeteGCNSetting=Inductive2021.09 | 93.89 | 0.45 | — | |
| TF-IDF + WideMLPSetting=Inductive2021.09 | 93.67 | 0.23 | — | |
| GloVe + WideMLPSetting=Inductive2021.09 | 93.58 | 0.06 | — | |
| TextGCNSetting=Transductive2021.09 | 93.56 | — | — | |
| TextGCN2026.04 | 93.56 | — | — | |
| TMPre-training=NTM with Top2vec2026.06 | 93.24 | — | — | |
| GloVe + WideMLP-2Setting=Inductive2021.09 | 93.19 | 0.11 | — | |
| SWEM2021.04 | 92.94 | — | — | |
| SWEMSetting=Inductive2021.09 | 92.94 | 0.24 | — | |
| fastText2021.04 | 92.81 | — | — | |
| fastTextSetting=Inductive2021.09 | 92.81 | 0.09 | — | |
| Drop Clause Tsetlin MachineDrop Clause Probability=0.752021.05 | 92.75 | — | — | |
| Graph-CNN-C2021.04 | 92.74 | — | — | |
| LEAM2021.04 | 91.84 | — | — | |
| fastText (bigrams)n-gram=bigrams2021.04 | 90.99 | — | — | |
| fastTexth=102026.06 | 90.99 | — | — | |
| Log. RegressionSetting=Inductive2021.09 | 90.65 | — | — | |
| Bi-LSTM2021.04 | 90.54 | — | — | |
| LSTM (pretrain)pre-training=true2021.04 | 90.48 | — | — | |
| LSTM (pretrain)Setting=Inductive2021.09 | 90.48 | 0.86 | — | |
| RNN/LSTMEmbedding=GloVe2026.04 | 90.48 | — | — | |
| LSTMEmbedding=GloVe2026.06 | 90.48 | — | — | |
| BERT2021.04 | 89.66 | — | — | |
| TextCNNPre-training=false, Training=true2026.06 | 89.5 | — | — | |
| TextGCNSetting=Inductive2021.09 | 89.39 | 0.38 | — | |
| TM with ϕfeature extension=cosine angle theta2021.04 | 89.14 | — | — | |
| Tsetlin MachineDrop Clause Probability=02021.05 | 89.14 | — | — | |
| TMEmbedding=GloVe2026.04 | 89.14 | — | — | |
| TMEmbedding=GloVe2026.06 | 89.14 | — | — | |
| TM with kfeature extension=k nearest words2021.04 | 88.59 | — | — | |
| CNN-non-staticstatic=false2021.04 | 87.59 | — | — | |
| TF-IDF+LR2021.04 | 86.95 | — | — | |
| BoW TFIDF2026.04 | 86.95 | — | — | |
| BoW TFIDF2026.06 | 86.95 | — | — | |
| Bag-of-WordsPre-training=false, Training=true2026.06 | 86.7 | — | — | |
| LDicePre-training=false, Training=false, k=72026.06 | 86.4 | — | — | |
| LJaccardPre-training=false, Training=false, k=72026.06 | 86.4 | — | — | |
| NCDlpPre-training=false, Training=false, k=72026.06 | 86.3 | — | — | |
| W2VPre-training=true, Training=false2026.06 | 85.6 | — | — | |
| LSTM2021.04 | 85.54 | — | — | |
| RNN/LSTM2026.04 | 85.54 | — | — | |
| LSTM2026.06 | 85.54 | — | — | |
| LSTMPre-training=false, Training=true2026.06 | 85.5 | — | — | |
| CNN-randinitialization=random2021.04 | 85.37 | — | — | |
| char-CNN2026.04 | 85.37 | — | — | |
| char-CNN2026.06 | 85.37 | — | — | |
| TMrepresentation=BOW2021.04 | 84.62 | — | — | |
| TM2026.04 | 84.62 | — | — | |
| TMDescription=Vanilla TM2026.06 | 84.62 | — | — | |
| NCDgzipPre-training=false, Training=false, k=72026.06 | 83.4 | — | — | |
| PV-DBOWarchitecture=Distributed Bag of Words2021.04 | 78.29 | — | — | |
| XGBoostLibrary=sklearn, Parameters=default2021.05 | 75.4 | — | — | |
| RFLibrary=sklearn, Parameters=default2021.05 | 72.5 | — | — | |
| SVMLibrary=sklearn, Parameters=default2021.05 | 71 | — | — | |
| K-NNLibrary=sklearn, Parameters=default2021.05 | 58.2 | — | — | |
| PV-DMarchitecture=Distributed Memory2021.04 | 44.92 | — | — | |
| TextLengthPre-training=false, Training=false2026.06 | 36.2 | — | — |