Text Classification on MR (test)
93.13AccuracyPLR-4
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| PLR-4Model=Llama3.1-8B-Instruct, k=322026.03 | 93.13 | — | — | — | — | |
| StaticModel=Llama3.1-8B-Instruct, k=322026.03 | 92.39 | — | — | — | — | |
| PLR-EMAModel=Qwen2.5-7B-Instruct, k=322026.03 | 92.21 | — | — | — | — | |
| PLR-1Model=Qwen2.5-7B-Instruct, k=322026.03 | 92.14 | — | — | — | — | |
| StaticModel=Llama3.1-8B-Instruct, k=162026.03 | 92.07 | — | — | — | — | |
| PLR-4Model=Qwen2.5-7B-Instruct, k=322026.03 | 92.02 | — | — | — | — | |
| Top-KModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.96 | — | — | — | — | |
| Top-KModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.95 | — | — | — | — | |
| PLR-1Model=Qwen2.5-7B-Instruct, k=162026.03 | 91.95 | — | — | — | — | |
| LocalEModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.9 | — | — | — | — | |
| PDO-UModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.87 | — | — | — | — | |
| PDO-UpModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.87 | — | — | — | — | |
| StaticModel=Llama3.1-8B-Instruct, k=82026.03 | 91.87 | — | — | — | — | |
| PDO-UModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.85 | — | — | — | — | |
| PDO-UpModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.83 | — | — | — | — | |
| PLR-EMAModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.83 | — | — | — | — | |
| GlobalEModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.8 | — | — | — | — | |
| PLR-4Model=Qwen2.5-7B-Instruct, k=162026.03 | 91.79 | — | — | — | — | |
| StaticModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.75 | — | — | — | — | |
| StaticModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.73 | — | — | — | — | |
| PDOModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.73 | — | — | — | — | |
| DeMOModel=Qwen2.5-7B-Instruct, k=322026.03 | 91.7 | — | — | — | — | |
| GlobalEModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.69 | — | — | — | — | |
| DeMOModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.67 | — | — | — | — | |
| PLR-EMAModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.46 | — | — | — | — | |
| PLR-4Model=Qwen2.5-7B-Instruct, k=82026.03 | 91.4 | — | — | — | — | |
| PDOModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.39 | — | — | — | — | |
| LocalEModel=Qwen2.5-7B-Instruct, k=162026.03 | 91.36 | — | — | — | — | |
| StaticModel=Llama3.1-8B-Instruct, k=42026.03 | 91.34 | — | — | — | — | |
| PLR-1Model=Qwen2.5-7B-Instruct, k=82026.03 | 91.32 | — | — | — | — | |
| PRL2025.12 | 91.27 | — | — | — | — | |
| LocalEModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.2 | — | — | — | — | |
| DeMOModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.17 | — | — | — | — | |
| Top-KModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.14 | — | — | — | — | |
| GlobalEModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.11 | — | — | — | — | |
| StaticModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.1 | — | — | — | — | |
| PDO-UpModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.1 | — | — | — | — | |
| PDOModel=Qwen2.5-7B-Instruct, k=82026.03 | 91.08 | — | — | — | — | |
| PIAST (E)variant=Ensemble2025.12 | 91 | — | — | — | — | |
| PDO-UModel=Qwen2.5-7B-Instruct, k=82026.03 | 90.91 | — | — | — | — | |
| Top-KModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.9 | — | — | — | — | |
| NI2025.12 | 90.85 | — | — | — | — | |
| PLR-4Model=Qwen2.5-7B-Instruct, k=42026.03 | 90.82 | — | — | — | — | |
| DP SimHash Bucketingprivacy budget (ε)=7, number of hyperplanes=24, number of hash tables=42026.05 | 90.8 | — | — | — | — | |
| DP SimHash Bucketingprivacy budget (ε)=8, number of hyperplanes=24, number of hash tables=42026.05 | 90.8 | — | — | — | — | |
| DeMOModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.79 | — | — | — | — | |
| PLR-1Model=Qwen2.5-7B-Instruct, k=42026.03 | 90.78 | — | — | — | — | |
| LocalEModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.72 | — | — | — | — | |
| DP SimHash Bucketingprivacy budget (ε)=5, number of hyperplanes=24, number of hash tables=42026.05 | 90.7 | — | — | — | — | |
| PLR-EMAModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.67 | — | — | — | — | |
| GlobalEModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.66 | — | — | — | — | |
| PDO-UModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.61 | — | — | — | — | |
| PDO-UpModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.61 | — | — | — | — | |
| DP SimHash Bucketingprivacy budget (ε)=3, number of hyperplanes=24, number of hash tables=42026.05 | 90.6 | — | — | — | — | |
| PIAST2025.12 | 90.57 | — | — | — | — | |
| PDOModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.56 | — | — | — | — | |
| DP SimHash Bucketingprivacy budget (ε)=1, number of hyperplanes=24, number of hash tables=42026.05 | 90.5 | — | — | — | — | |
| GA2025.12 | 90.45 | — | — | — | — | |
| PIAST (I)variant=Iterative2025.12 | 90.43 | — | — | — | — | |
| PIAST (LOO)variant=Leave-one-out2025.12 | 90.42 | — | — | — | — | |
| StaticModel=Qwen2.5-7B-Instruct, k=42026.03 | 90.22 | — | — | — | — | |
| DE2025.12 | 89.98 | — | — | — | — | |
| APO2025.12 | 89.97 | — | — | — | — | |
| APE2025.12 | 89.9 | — | — | — | — | |
| ROBERTAGCNBackbone=RoBERTa2021.05 | 89.7 | — | — | — | — | |
| KNN-Promptingprivacy budget (ε)=Non-private, number of hyperplanes=24, number of hash tables=42026.05 | 89.7 | — | — | — | — | |
| RoBERTa2021.05 | 89.4 | — | — | — | — | |
| ROBERTAGATBackbone=RoBERTa2021.05 | 89.2 | — | — | — | — | |
| DP SimHash Bucketingprivacy budget (ε)=0.1, number of hyperplanes=24, number of hash tables=42026.05 | 88.4 | — | — | — | — | |
| Fine-tuned BERTSNN=false, Spike=false, PE=false, Param (M)=109.82024.05 | 87.63 | — | — | — | — | |
| MI2025.12 | 87.4 | — | — | — | — | |
| PINFew-shot=5-shot, Policy LM=OPT-125M2024.07 | 87.2 | — | — | — | — | |
| RLPromptFew-shot=5-shot, Policy LM=OPT-125M2024.07 | 87.1 | — | — | — | — | |
| BYTE MLSTMClassifier=Logistic Regression, Penalty=L12017.04 | 86.9 | — | — | — | — | |
| BertGATBackbone=BERT-base2021.05 | 86.5 | — | — | — | — | |
| GrIPSFew-shot=5-shot2024.07 | 86.1 | — | — | — | — | |
| BertGCNBackbone=BERT-base2021.05 | 86 | — | — | — | — | |
| BERTModel variant=BERT-base2021.05 | 85.7 | — | — | — | — | |
| InstructionsFew-shot=5-shot2024.07 | 85.2 | — | — | — | — | |
| ADASENT2017.04 | 83.1 | — | — | — | — | |
| Zhao et al. (2015)2017.09 | 83.1 | — | — | — | — | |
| SRU (8 layers)Size=502k, Optimizer=Adam, Word embeddings=fixed, Evaluation protocol=3x10 runs 10-fold CV2017.09 | 83.1 | — | — | — | — | |
| SRU (4 layers)Size=303k, Optimizer=Adam, Word embeddings=fixed, Evaluation protocol=3x10 runs 10-fold CV2017.09 | 82.9 | — | — | — | — | |
| Spikformer w/ CPG-PESNN=true, Spike=true, PE=true, Param (M)=110.42024.05 | 82.42 | — | — | — | — | |
| SRU (2 layers)Size=204k, Optimizer=Adam, Word embeddings=fixed, Evaluation protocol=3x10 runs 10-fold CV2017.09 | 82.3 | — | — | — | — | |
| QRNN (k=1) + highwaySize=204k, Optimizer=Adam, Word embeddings=fixed, Evaluation protocol=3x10 runs 10-fold CV2017.09 | 82.1 | — | — | — | — | |
| QRNN (k=1)Size=165k, Optimizer=Adam, Word embeddings=fixed, Evaluation protocol=3x10 runs 10-fold CV2017.09 | 82 | — | — | — | — | |
| Zhang and Wallace (2017)2017.09 | 81.7 | — | — | — | — | |
| CNN2017.04 | 81.5 | — | — | — | — | |
| CNN2018.05 | 81.5 | — | — | — | — | |
| Kim (2014)2017.09 | 81.5 | — | — | — | — | |
| TextGSL2025.12 | 81.27 | — | — | — | — | |
| Manual PromptFew-shot=5-shot2024.07 | 80.9 | — | — | — | — | |
| In-Context DemonstrationFew-shot=5-shot2024.07 | 80.6 | — | — | — | — | |
| LSTMSize=352k, Optimizer=Adam, Word embeddings=fixed, Evaluation protocol=3x10 runs 10-fold CV2017.09 | 79.8 | — | — | — | — | |
| TextSSL2025.12 | 79.74 | — | — | — | — | |
| Spikformer w/ Float-PESNN=true, Spike=false, PE=true, Param (M)=109.82024.05 | 79.67 | — | — | — | — | |
| SKIPTHOUGHTLayer Normalization=True2017.04 | 79.5 | — | — | — | — | |
| NBSVM2017.04 | 79.4 | — | — | — | — | |
| UGformerModel Variant=Variant 2, GNN layer usage=with GNN2019.09 | 79.29 | — | — | — | — |