Sentiment Analysis on SST-5
94.84AccuracyMUPPET
Evaluation Results
| Method | Links | |
|---|---|---|
| MUPPET2022.12 | 94.84 | |
| ColD-Fusion2022.12 | 59.52 | |
| ROBERTa-large+Self-ExplainingBackbone=RoBERTa-large2020.12 | 59.1 | |
| SentiLAREBackbone=RoBERTa-base, External training resources=true2020.12 | 58.6 | |
| ROBERTa-largeBackbone=RoBERTa-large2020.12 | 57.9 | |
| ROBERTa-base+Self-ExplainingBackbone=RoBERTa-base2020.12 | 57.8 | |
| Multitask2022.12 | 57.56 | |
| SentiBERT2020.12 | 56.9 | |
| ROBERTa-baseBackbone=RoBERTa-base2020.12 | 56.4 | |
| Finetune2022.12 | 56.24 | |
| BCN+SuBiLSTM+CoVeBackbone=SuBiLSTM+CoVe2020.12 | 56.2 | |
| BERT-largeBackbone=BERT-large2020.12 | 56.2 | |
| ROBERTa-base+AvgSelf-ExplainingBackbone=RoBERTa-base2020.12 | 56.2 | |
| BM25Model=Qwen2.5-7B, Setting=Few-shot2026.05 | 55.6 | |
| PCModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 55.41 | |
| BERT-baseBackbone=BERT-base2020.12 | 54.9 | |
| CCBackbone Model=Mistral, Shot Count=16-shot2025.05 | 54.3 | |
| CCModel=Mistral, k-shot=162025.05 | 54.3 | |
| DiSPModel=LLaMA3-8B, Setting=Few-shot2026.05 | 54.3 | |
| UpriseModel=Qwen2.5-7B, Setting=Few-shot2026.05 | 54.3 | |
| DiSPModel=Qwen2.5-7B, Setting=Few-shot2026.05 | 54.3 | |
| DCBackbone Model=Mistral, Shot Count=16-shot2025.05 | 54.14 | |
| DCModel=Mistral, k-shot=162025.05 | 54.14 | |
| UniBiasModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 53.79 | |
| PCModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 53.63 | |
| Se2Model=Qwen2.5-7B, Setting=Few-shot2026.05 | 53.6 | |
| Se²2024.02 | 52.7 | |
| UPRISE2024.02 | 52.6 | |
| CCBackbone Model=Llama, Shot Count=16-shot2025.05 | 52.58 | |
| CCModel=Llama, k-shot=162025.05 | 52.58 | |
| BM25Model=LLaMA3-8B, Setting=Few-shot2026.05 | 52.3 | |
| DCModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 51.92 | |
| gold shot2026.02 | 51.9 | |
| UniBiasModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 51.8 | |
| Se2Model=LLaMA3-8B, Setting=Few-shot2026.05 | 51.7 | |
| UCPOF2026.02 | 51.54 | |
| baseline2026.02 | 51.45 | |
| Full RAG2026.02 | 51.4 | |
| CCModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 51.36 | |
| ICLModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 51.03 | |
| DCBackbone Model=Llama, Shot Count=16-shot2025.05 | 50.7 | |
| DCModel=Llama, k-shot=162025.05 | 50.7 | |
| Full RAGModel=Llama-3-8B-Instruct2026.02 | 50.63 | |
| LOZO-MBackbone=RoBERTa-Large, Evaluation Protocol=prompt-conditioned, Time Constraint=equal wall-clock time2025.11 | 50.4 | |
| ConMeZOBackbone=RoBERTa-Large, Evaluation Protocol=prompt-conditioned, Time Constraint=equal wall-clock time2025.11 | 50 | |
| RandomModel=Qwen2.5-7B, Setting=Few-shot2026.05 | 49.7 | |
| LOZOBackbone=RoBERTa-Large, Evaluation Protocol=prompt-conditioned, Time Constraint=equal wall-clock time2025.11 | 49.4 | |
| DCModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 48.98 | |
| CLGModel=Llama3-70B, Shots=1282025.06 | 48.32 | |
| UpriseModel=LLaMA3-8B, Setting=Few-shot2026.05 | 48.3 | |
| baselineModel=Llama-3-8B-Instruct2026.02 | 48.1 | |
| SCBackbone Model=Mistral, Shot Count=16-shot2025.05 | 47.27 | |
| SCModel=Mistral, k-shot=162025.05 | 47.27 | |
| CCModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 47.2 | |
| UCPOFModel=Llama-3-8B-Instruct2026.02 | 47.19 | |
| SeDPOModel=LLaMA3-8B, Setting=Few-shot2026.05 | 47 | |
| ICLModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 46.97 | |
| RandomModel=LLaMA3-8B, Setting=Few-shot2026.05 | 46.6 | |
| Latent-BayesianModel=Llama3-70B, Shots=1282025.06 | 46.59 | |
| gold shotModel=Llama-3-8B-Instruct2026.02 | 46.52 | |
| BGE-KMeansModel=Llama3-70B, Shots=1282025.06 | 46.41 | |
| SeDPOModel=Qwen2.5-7B, Setting=Few-shot2026.05 | 45.7 | |
| SCBackbone Model=Llama, Shot Count=16-shot2025.05 | 44.45 | |
| SCModel=Llama, k-shot=162025.05 | 44.45 | |
| RandomModel=Llama3-70B, Shots=1282025.06 | 43.71 | |
| BM25-MajorModel=Llama3-70B, Shots=1282025.06 | 43.69 | |
| Best-of-NModel=Llama3-70B, Shots=1282025.06 | 43.69 | |
| EPR-KMeansModel=Llama3-70B, Shots=1282025.06 | 43.05 | |
| Zero-shotModel=Qwen2.5-7B, Setting=Zero-shot2026.05 | 42.4 | |
| Zero-shotModel=LLaMA3-8B, Setting=Zero-shot2026.05 | 41.7 | |
| SCBackbone Model=Qwen, Shot Count=16-shot2025.05 | 41.64 | |
| SCModel=Qwen, k-shot=162025.05 | 41.64 | |
| CLGModel=Qwen2.5-72B, Shots=1282025.06 | 38.33 | |
| BM25-MajorModel=Qwen2.5-72B, Shots=1282025.06 | 37.78 | |
| BGE-KMeansModel=Qwen2.5-72B, Shots=1282025.06 | 37.24 | |
| Best-of-NModel=Qwen2.5-72B, Shots=1282025.06 | 36.6 | |
| EPR-KMeansModel=Qwen2.5-72B, Shots=1282025.06 | 36.15 | |
| RandomModel=Qwen2.5-72B, Shots=1282025.06 | 36 | |
| RAEE (Llama)Backbone=Llama-3-8B2024.05 | 35.25 | |
| RB-LBackbone=RoBERTa-Large2024.05 | 34.98 | |
| BCBackbone Model=Mistral, Shot Count=16-shot2025.05 | 34.69 | |
| BCModel=Mistral, k-shot=162025.05 | 34.69 | |
| RAEE (RB-L)Backbone=RoBERTa-Large2024.05 | 33.57 | |
| BCBackbone Model=Llama, Shot Count=16-shot2025.05 | 33.36 | |
| BCModel=Llama, k-shot=162025.05 | 33.36 | |
| RAEE (Gemma)Backbone=Gemma-7B2024.05 | 32.4 | |
| Gemma-7BBackbone=Gemma-7B2024.05 | 28.64 | |
| Zero-shot2024.02 | 28.2 | |
| AdaInfer (T5-L)Backbone=T5-Large2024.05 | 28.14 | |
| Best-of-102024.02 | 27.7 | |
| BM252024.02 | 27.5 | |
| Latent-BayesianModel=Qwen2.5-72B, Shots=1282025.06 | 27.25 | |
| RAEE (T5-L)Backbone=T5-Large2024.05 | 26.56 | |
| Llama-3-8BBackbone=Llama-3-8B2024.05 | 26.06 | |
| AdaInfer (RB-L)Backbone=RoBERTa-Large2024.05 | 24.48 | |
| CALM (T5-L)Backbone=T5-Large2024.05 | 23.17 | |
| T5-LBackbone=T5-Large2024.05 | 23.12 | |
| BCBackbone Model=Qwen, Shot Count=16-shot2025.05 | 22.89 | |
| BCModel=Qwen, k-shot=162025.05 | 22.89 | |
| Base LLMBackbone Model=Qwen, Shot Count=16-shot2025.05 | 22.81 |