Text Classification on SST-2
96.32AccuracyPRL
Evaluation Results
| Method | Links | |
|---|---|---|
| PRLBackbone=Qwen2.5-7B-Instruct, Prompt Selection=True2025.05 | 96.32 | |
| Heinsen RoutingBackbone=RoBERTa-large, Transformer Status=Frozen, Routing Iterations (n_iters)=22022.11 | 96 | |
| PRLBackbone=Qwen2.5-7B-Instruct, Prompt Selection=False2025.05 | 95.98 | |
| NIBackbone=Qwen2.5-7B-Instruct2025.05 | 95.77 | |
| FPFTBackbone=Llama2, k (training examples per class)=200, #Param=6.7B2024.02 | 95.64 | |
| LORABackbone=Llama2, k (training examples per class)=5, #Param=4.2M2024.02 | 95.42 | |
| GNNAVI-GCNBackbone=Llama2, k (training examples per class)=200, #Param=16.8M2024.02 | 95.36 | |
| GNNAVI-SAGEBackbone=Llama2, k (training examples per class)=200, #Param=33.6M2024.02 | 95.3 | |
| SubRegWeigh (K-means)Backbone=RoBERTa-LARGE, Processing time (hh:mm)=5:212024.09 | 94.84 | |
| SubRegWeigh (Cos-Sim)Backbone=RoBERTa-LARGE, Processing time (hh:mm)=4:512024.09 | 94.75 | |
| VanillaBackbone=RoBERTa-LARGE2024.09 | 94.68 | |
| GABackbone=Qwen2.5-7B-Instruct2025.05 | 94.65 | |
| FPFTBackbone=Llama2, k (training examples per class)=5, #Param=6.7B2024.02 | 94.63 | |
| SubRegWeigh (Random)Backbone=RoBERTa-LARGE, Processing time (hh:mm)=3:262024.09 | 94.61 | |
| PromptAgentBackbone=Qwen2.5-7B-Instruct2025.05 | 94.58 | |
| Res-TuningTrain Time=7.9, Train Param.=0.97 (0.77%), Test Param.=0.97 (0.77%), Mem.=19.3G2023.10 | 94.56 | |
| GNNAVI-GCNBackbone=Llama2, k (training examples per class)=5, #Param=16.8M2024.02 | 94.56 | |
| CrossWeighBackbone=RoBERTa-LARGE, Processing time (hh:mm)=30:552024.09 | 94.31 | |
| MAM AdapterTrain Time=7.2, Train Param.=46.78 (37.4%), Test Param.=0.61 (0.5%), Mem.=22.4G2023.10 | 94.2 | |
| PromptWizardBackbone=Qwen2.5-7B-Instruct2025.05 | 93.87 | |
| APOBackbone=Qwen2.5-7B-Instruct2025.05 | 93.71 | |
| GRACEBackbone=Qwen2.5-7B-Instruct2025.05 | 93.61 | |
| DEBackbone=Qwen2.5-7B-Instruct2025.05 | 93.29 | |
| Res-Tuning-BypassTrain Time=4.2, Train Param.=0.98 (0.78%), Test Param.=0.98 (0.78%), Mem.=4.3G2023.10 | 92.94 | |
| GNNAVI-SAGEBackbone=Llama2, k (training examples per class)=5, #Param=33.6M2024.02 | 92.91 | |
| MIBackbone=Qwen2.5-7B-Instruct2025.05 | 92.7 | |
| LORABackbone=GPT2-XL, k (training examples per class)=5, #Param=2.5M2024.02 | 91.98 | |
| PITOMEType=merging, Eval Flops=x1.9, Train Speed=x1.42024.05 | 91.7 | |
| BERTType=baseline, Eval Flops=x1.0, Train Speed=x1.02024.05 | 91.4 | |
| ALBERTType=compressed models, Eval Flops=x1.0, Train Speed=x1.12024.05 | 91.3 | |
| FisherType=pruning + mask, Eval Flops=x1.6, Train Speed=x1.02024.05 | 91.3 | |
| LTPType=pruning + mask, Eval Flops=x2.9, Train Speed=x1.02024.05 | 91.3 | |
| LORABackbone=Llama2, k (training examples per class)=200, #Param=4.2M2024.02 | 91.29 | |
| APEBackbone=Qwen2.5-7B-Instruct2025.05 | 91.23 | |
| ToMeType=merging, Eval Flops=x1.9, Train Speed=x1.42024.05 | 91.2 | |
| DistilBERTType=compressed models, Eval Flops=x2.0, Train Speed=x1.72024.05 | 91.1 | |
| PowER-BERTType=pruning + mask, Eval Flops=x2.5, Train Speed=x1.02024.05 | 91.1 | |
| PITOMEType=pruning + mask, Eval Flops=x1.9, Train Speed=x1.42024.05 | 91 | |
| LORABackbone=GPT2-XL, k (training examples per class)=200, #Param=2.5M2024.02 | 90.83 | |
| DCTType=merging, Eval Flops=x1.9, Train Speed=x1.42024.05 | 90.7 | |
| GNNAVI-GCNBackbone=GPT2-XL, k (training examples per class)=200, #Param=2.6M2024.02 | 90.67 | |
| Original2022.07 | 90.5 | |
| GNNAVI-SAGEBackbone=GPT2-XL, k (training examples per class)=200, #Param=5.1M2024.02 | 90.46 | |
| Latent Concept LearningBackbone=GPT2-large, Configuration=concept tokens as prefixes2023.01 | 90.3 | |
| ToFuType=merging, Eval Flops=x1.9, Train Speed=x1.42024.05 | 89.8 | |
| DiffRateType=merging, Eval Flops=x1.9, Train Speed=x1.42024.05 | 89.7 | |
| BLSTM-2DCNNNN category=ours2016.11 | 89.5 | |
| MVCNNNN category=CNN2016.11 | 89.4 | |
| DSCNNNN category=Other2016.11 | 89.1 | |
| AdapterBackbone=GPT2-XL, k (training examples per class)=200, #Param=15.4M2024.02 | 88.65 | |
| Molding-CNNNN category=CNN2016.11 | 88.6 | |
| BLSTM-2DPoolingNN category=ours2016.11 | 88.3 | |
| MPAD-sentence-att2019.08 | 88.3 | |
| BLSTM-AttNN category=ours2016.11 | 88.2 | |
| CNN-MCNN category=CNN2016.11 | 88.1 | |
| Tree-LSTMNN category=RNN2016.11 | 88 | |
| LSTM-RNNNN category=Other2016.11 | 88 | |
| Tree-LSTM2019.08 | 88 | |
| MPAD-clique2019.08 | 87.91 | |
| Shared-Layer ArchitectureMulti-Task=true, Fine-Tuning=true2016.05 | 87.9 | |
| TBCNNNN category=CNN2016.11 | 87.9 | |
| Multi-TaskNN category=RNN2016.11 | 87.9 | |
| PVNN category=Other2016.11 | 87.8 | |
| C-LSTMNN category=Other2016.11 | 87.8 | |
| doc2vec2019.08 | 87.8 | |
| C-LSTM2019.08 | 87.8 | |
| MPAD2019.08 | 87.8 | |
| MPAD-path2019.08 | 87.75 | |
| BLSTMNN category=ours2016.11 | 87.6 | |
| BLSTMNN category=RNN2016.11 | 87.5 | |
| LR-LSTM2018.03 | 87.5 | |
| LSTMNNN category=RNN2016.11 | 87.3 | |
| CNN-non-staticNN category=CNN2016.11 | 87.2 | |
| CNNmode=non-static2018.03 | 87.2 | |
| CNN2019.08 | 87.2 | |
| LSTMN2019.08 | 87 | |
| ICLBackbone=Llama2, k (training examples per class)=52024.02 | 86.93 | |
| Tree-LSTM2016.05 | 86.9 | |
| DCNN2016.05 | 86.8 | |
| DCNNNN category=CNN2016.11 | 86.8 | |
| DANNN category=Other2016.11 | 86.8 | |
| CNNmode=static2018.03 | 86.8 | |
| Capsule-B2018.03 | 86.8 | |
| DiSAN2019.08 | 86.76 | |
| HN-ATT2019.08 | 86.71 | |
| DRNNNN category=ReNN2016.11 | 86.6 | |
| DRNN2019.08 | 86.6 | |
| Capsule-A2018.03 | 86.4 | |
| LSTM-GRNN2019.08 | 86.38 | |
| DAN2019.08 | 86.3 | |
| Tree-LSTM2018.03 | 85.7 | |
| CNN-AnaNN category=CNN2016.11 | 85.45 | |
| RNTN2016.05 | 85.4 | |
| RNTNNN category=ReNN2016.11 | 85.4 | |
| LSTMNN category=RNN2016.11 | 84.9 | |
| GNNAVI-GCNBackbone=GPT2-XL, k (training examples per class)=5, #Param=2.6M2024.02 | 84.31 | |
| BILSTM2018.03 | 83.2 | |
| MV-RNN2016.05 | 82.9 | |
| PV2016.05 | 82.7 | |
| CNNmode=rand2018.03 | 82.7 |