Sentiment Analysis on CR
96.4AccuracyFADS-ICL
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| FADS-ICLBackbone=Llama-1 30B, Shots=128-shots2024.05 | 96.4 | — | — | — | |
| DARTBackbone=RoBERTa-large, Training size=Full training set2021.08 | 93.8 | — | — | — | |
| FADS-ICLBackbone=Llama-2 13B, Shots=128-shots2024.05 | 93.8 | — | — | — | |
| kNN-promptingBackbone=Llama-2 7B, Shots=128-shots2024.05 | 93.7 | — | — | — | |
| kNN-promptBackbone=Llama-2 7B, Shots=128-shots2024.05 | 93.4 | — | — | — | |
| FADS-ICLBackbone=Llama-2 7B, Shots=128-shots2024.05 | 93.4 | — | — | — | |
| FADS-ICLBackbone=Llama-2 70B, Shots=128-shots2024.05 | 93.3 | — | — | — | |
| FADS-ICLBackbone=Llama-1 7B, Shots=128-shots2024.05 | 93.2 | — | — | — | |
| DCModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 92.61 | — | — | — | |
| UniBiasModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 92.61 | — | — | — | |
| kNN-promptingBackbone=Llama-1 30B, Shots=128-shots2024.05 | 92.6 | — | — | — | |
| kNN-promptBackbone=Llama-2 70B, Shots=128-shots2024.05 | 92.5 | — | — | — | |
| LM-BFFBackbone=RoBERTa-large, Training size=Full training set, Demonstration usage=without demonstration2021.08 | 92.4 | — | — | — | |
| UniBiasModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 92.34 | — | — | — | |
| kNN-promptBackbone=Llama-1 7B, Shots=128-shots2024.05 | 92.3 | — | — | — | |
| kNN-promptingBackbone=Llama-1 13B, Shots=128-shots2024.05 | 92.3 | — | — | — | |
| FADS-ICLBackbone=Llama-1 13B, Shots=128-shots2024.05 | 92.3 | — | — | — | |
| ICLBackbone=Llama-2 70B, Shots=128-shots2024.05 | 92.3 | — | — | — | |
| FADS-ICLcandidate pool size (m)=322024.05 | 92.3 | — | — | — | |
| DCModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 92.23 | — | — | — | |
| kNN-promptingBackbone=Llama-2 13B, Shots=128-shots2024.05 | 92.2 | — | — | — | |
| CCModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 92.13 | — | — | — | |
| CCModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 92.13 | — | — | — | |
| kNN-promptingBackbone=Llama-2 70B, Shots=128-shots2024.05 | 92 | — | — | — | |
| PCModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 91.97 | — | — | — | |
| kNN-promptingBackbone=Llama-1 7B, Shots=128-shots2024.05 | 91.9 | — | — | — | |
| PCModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 91.65 | — | — | — | |
| ICLModel=Llama-2 7b, ICL Setting=1-shot2024.05 | 91.54 | — | — | — | |
| kNN-promptBackbone=Llama-1 13B, Shots=128-shots2024.05 | 91.5 | — | — | — | |
| kNN-promptBackbone=Llama-2 13B, Shots=128-shots2024.05 | 91.5 | — | — | — | |
| kNN-promptBackbone=Llama-1 30B, Shots=128-shots2024.05 | 91.5 | — | — | — | |
| MPD2024.06 | 91.3 | — | — | 0.3 | |
| FADS-ICLNumber of training shots (m)=64, LLM scale=1.5B2024.05 | 91.2 | — | — | — | |
| ICLBackbone=Llama-1 30B, Shots=128-shots2024.05 | 91.2 | — | — | — | |
| FADS-ICLcandidate pool size (m)=642024.05 | 91.2 | — | — | — | |
| TEMPERA2024.06 | 91.1 | — | — | 1.6 | |
| ICLModel=Llama-2 13b, ICL Setting=1-shot2024.05 | 91.01 | — | — | — | |
| ROBERTA-12Dim=768, Fine-tuning status=true2021.06 | 90.8 | 90 | — | — | |
| DP2O2024.06 | 90.8 | — | — | 0.5 | |
| BERT-12Dim=768, Fine-tuning status=true2021.06 | 90.6 | 89.6 | — | — | |
| FADS-ICLNumber of training shots (m)=128, LLM scale=1.5B2024.05 | 90.5 | — | — | — | |
| FADS-ICLBackbone=GPT-2 1.5B, Shots=128-shots2024.05 | 90.5 | — | — | — | |
| FADS-ICLcandidate pool size (m)=1282024.05 | 90.5 | — | — | — | |
| ICLBackbone=Llama-2 7B, Shots=128-shots2024.05 | 90.4 | — | — | — | |
| bitfit+mteMode=Ablation2022.04 | 90.1 | — | 87.8 | 1 | |
| PERFECTInitialization=random2022.04 | 90 | — | 85.5 | 1.4 | |
| Logan IV et al. (2021)2022.04 | 89.9 | — | 87.2 | 1.1 | |
| ICLBackbone=Llama-1 7B, Shots=128-shots2024.05 | 89.8 | — | — | — | |
| PERFECTInitialization=prototype-based2022.04 | 89.7 | — | 87.4 | 1.2 | |
| FADS-ICLcandidate pool size (m)=2562024.05 | 89.6 | — | — | — | |
| GPT2-12Dim=768, Fine-tuning status=true2021.06 | 89.5 | 88.5 | — | — | |
| Fine-tuningBackbone=RoBERTa-large, Training size=Full training set2021.08 | 89.4 | — | — | — | |
| kNN-promptingNumber of training shots (m)=64, LLM scale=1.5B2024.05 | 89.1 | — | — | — | |
| PETStrategy=Best2022.04 | 88.8 | — | 85.8 | 1.9 | |
| kNN-promptingNumber of training shots (m)=128, LLM scale=1.5B2024.05 | 88.8 | — | — | — | |
| kNN-promptingBackbone=GPT-2 1.5B, Shots=128-shots2024.05 | 88.8 | — | — | — | |
| SBERT-BASE-NLIDim=768, Fine-tuning status=false2021.06 | 88.7 | 87.4 | — | — | |
| FADS-ICLNumber of training shots (m)=16, LLM scale=1.5B2024.05 | 88.7 | — | — | — | |
| FADS-ICLBackbone=GPT-2 0.8B, Shots=128-shots2024.05 | 88.7 | — | — | — | |
| SROBERTA-BASE-NLIDim=768, Fine-tuning status=false2021.06 | 88.6 | 87.6 | — | — | |
| PETStrategy=Average2022.04 | 88.4 | — | 68.8 | 3 | |
| FADS-ICLNumber of training shots (m)=8, LLM scale=1.5B2024.05 | 88 | — | — | — | |
| BERT Large FTBackbone=BERT Large, Shots=N/A (Fine-tuned)2024.05 | 88 | — | — | — | |
| ICLBackbone=Llama-2 13B, Shots=128-shots2024.05 | 87.8 | — | — | — | |
| kNN-promptNumber of training shots (m)=128, LLM scale=1.5B2024.05 | 87 | — | — | — | |
| kNN-promptBackbone=GPT-2 1.5B, Shots=128-shots2024.05 | 87 | — | — | — | |
| GPT Large FTBackbone=GPT Large, Shots=N/A (Fine-tuned)2024.05 | 86.7 | — | — | — | |
| kNN-promptingBackbone=GPT-2 0.8B, Shots=128-shots2024.05 | 86.6 | — | — | — | |
| kNN-promptNumber of training shots (m)=64, LLM scale=1.5B2024.05 | 86.5 | — | — | — | |
| kNN-promptBackbone=GPT-2 0.8B, Shots=128-shots2024.05 | 86 | — | — | — | |
| kNN-promptingNumber of training shots (m)=32, LLM scale=1.5B2024.05 | 85.3 | — | — | — | |
| kNN-promptNumber of training shots (m)=32, LLM scale=1.5B2024.05 | 85.2 | — | — | — | |
| kNN-promptingNumber of training shots (m)=4, LLM scale=1.5B2024.05 | 84.9 | — | — | — | |
| kNN-promptNumber of training shots (m)=8, LLM scale=1.5B2024.05 | 84.5 | — | — | — | |
| BERT2STATICparaDim=768, Fine-tuning status=false2021.06 | 83.6 | 81.1 | — | — | |
| GPT22STATICsentDim=768, Fine-tuning status=false2021.06 | 83.5 | 81.5 | — | — | |
| kNN-promptingNumber of training shots (m)=8, LLM scale=1.5B2024.05 | 83.5 | — | — | — | |
| kNN-promptNumber of training shots (m)=4, LLM scale=1.5B2024.05 | 83.3 | — | — | — | |
| ICLBackbone=GPT-2 0.8B, Shots=128-shots2024.05 | 83.3 | — | — | — | |
| GLOVEDim=300, Fine-tuning status=false2021.06 | 83.2 | 81.6 | — | — | |
| ASE - BERT-12Best Layer=5, Dim=768, Fine-tuning status=false2021.06 | 83 | 81.5 | — | — | |
| kNN-promptingNumber of training shots (m)=16, LLM scale=1.5B2024.05 | 82.7 | — | — | — | |
| GPT22STATICparaDim=768, Fine-tuning status=false2021.06 | 82.6 | 81 | — | — | |
| ROBERTA2STATICparaDim=768, Fine-tuning status=false2021.06 | 82.3 | 80.9 | — | — | |
| kNN-promptNumber of training shots (m)=16, LLM scale=1.5B2024.05 | 82.3 | — | — | — | |
| ASE - GPT2-12Best Layer=4, Dim=768, Fine-tuning status=false2021.06 | 82.1 | 81 | — | — | |
| BERT2STATICsentDim=768, Fine-tuning status=false2021.06 | 82 | 80.1 | — | — | |
| FASTTEXTDim=300, Fine-tuning status=false2021.06 | 81.9 | 80.3 | — | — | |
| ICLBackbone=Llama-1 13B, Shots=128-shots2024.05 | 81.5 | — | — | — | |
| SENT2VECDim=768, Fine-tuning status=false2021.06 | 81.2 | 79.8 | — | — | |
| ASE - ROBERTA-12Best Layer=2, Dim=768, Fine-tuning status=false2021.06 | 81.2 | 78.4 | — | — | |
| GLOVE (Twitter)Dim=200, Fine-tuning status=false2021.06 | 80.9 | 79 | — | — | |
| SKIPGRAMDim=768, Fine-tuning status=false2021.06 | 80.9 | 78.4 | — | — | |
| FINETUNE2022.04 | 80.1 | — | 72.9 | 4.1 | |
| ROBERTA2STATICsentDim=768, Fine-tuning status=false2021.06 | 79.6 | 76.5 | — | — | |
| RB-LBackbone=RoBERTa-Large2024.05 | 79.55 | — | — | — | |
| CBOWDim=768, Fine-tuning status=false2021.06 | 78.5 | 75.9 | — | — | |
| Llama-3-8BBackbone=Llama-3-8B2024.05 | 72.9 | — | — | — | |
| prompt+mteMode=Ablation2022.04 | 71 | — | 55.8 | 8.2 | |
| RAEE (RB-L)Backbone=RoBERTa-Large2024.05 | 68.05 | — | — | — |