List your company
ResearchFormerly WizWand

Sentiment Analysis on MR

0.939AccuracyICL
0.823560.853530.88350.91347Jun 8, 2021Dec 6, 2021Jun 5, 2022Dec 4, 2022Jun 3, 2023Dec 1, 2023May 31, 2024
Updated 8mo ago

Evaluation results

MethodLinks
ICLBackbone=Llama-2 70B, Shots=128-shots2024.050.939—
FADS-ICLBackbone=Llama-1 30B, Shots=128-shots2024.050.938—
FADS-ICLBackbone=Llama-1 13B, Shots=128-shots2024.050.934—
FADS-ICLBackbone=Llama-2 13B, Shots=128-shots2024.050.934—
RLLRBackbone=Mistral 7B2024.050.933—
kNN-promptBackbone=Llama-2 7B, Shots=128-shots2024.050.932—
CCModel=Llama-2 13b, ICL Setting=1-shot2024.050.932—
kNN-promptingBackbone=Llama-1 13B, Shots=128-shots2024.050.931—
FADS-ICLBackbone=Llama-2 70B, Shots=128-shots2024.050.93—
DCModel=Llama-2 13b, ICL Setting=1-shot2024.050.93—
ICLBackbone=Llama-2 7B, Shots=128-shots2024.050.929—
PCModel=Llama-2 13b, ICL Setting=1-shot2024.050.928—
kNN-promptBackbone=Llama-1 7B, Shots=128-shots2024.050.927—
kNN-promptBackbone=Llama-1 13B, Shots=128-shots2024.050.927—
kNN-promptBackbone=Llama-2 13B, Shots=128-shots2024.050.927—
RLLR_MIXEDBackbone=LLaMA2 7B2024.050.926—
RLLRMIXEDModel=LLaMA2, Model Size=7B2024.050.926—
RLLRModel=LLaMA2, Model Size=13B2024.050.926—
RLLRBackbone=LLaMA2 7B2024.050.924—
RLLR_MIXEDBackbone=Mistral 7B2024.050.924—
RLLRModel=LLaMA2, Model Size=7B2024.050.924—
RLLRMIXEDModel=LLaMA2, Model Size=13B2024.050.924—
FADS-ICLBackbone=Llama-1 7B, Shots=128-shots2024.050.924—
DCModel=Llama-2 7b, ICL Setting=1-shot2024.050.9235—
UniBiasModel=Llama-2 13b, ICL Setting=1-shot2024.050.9223—
SFT w. rat.Model=LLaMA2, Model Size=13B2024.050.922—
RLHFModel=LLaMA2, Model Size=13B2024.050.922—
kNN-promptingBackbone=Llama-2 7B, Shots=128-shots2024.050.922—
FADS-ICLBackbone=Llama-2 7B, Shots=128-shots2024.050.922—
kNN-promptingBackbone=Llama-2 13B, Shots=128-shots2024.050.922—
kNN-promptingBackbone=Llama-1 30B, Shots=128-shots2024.050.922—
UniBiasModel=Llama-2 7b, ICL Setting=1-shot2024.050.9219—
SFTBackbone=Mistral 7B2024.050.921—
RLHFBackbone=Mistral 7B2024.050.921—
SFT w. rat.Backbone=Mistral 7B2024.050.92—
SFTModel=LLaMA2, Model Size=13B2024.050.92—
SFT w. rat.Backbone=LLaMA2 7B2024.050.919—
RLHFBackbone=LLaMA2 7B2024.050.919—
SFT w. rat.Model=LLaMA2, Model Size=7B2024.050.919—
RLHFModel=LLaMA2, Model Size=7B2024.050.919—
kNN-promptingBackbone=Llama-1 7B, Shots=128-shots2024.050.918—
CCModel=Llama-2 7b, ICL Setting=1-shot2024.050.9177—
kNN-promptBackbone=Llama-2 70B, Shots=128-shots2024.050.917—
kNN-promptingBackbone=Llama-2 70B, Shots=128-shots2024.050.916—
kNN-promptBackbone=Llama-1 30B, Shots=128-shots2024.050.915—
RLLRBackbone=ChatGLM3 6B2024.050.914—
RLLR_MIXEDBackbone=ChatGLM3 6B2024.050.914—
PCModel=Llama-2 7b, ICL Setting=1-shot2024.050.9139—
SFT w. rat.Backbone=ChatGLM3 6B2024.050.913—
RLLRBackbone=Baichuan2 7B2024.050.913—
RLHFBackbone=Baichuan2 7B2024.050.912—
RLLR_MIXEDBackbone=Baichuan2 7B2024.050.912—
ICLBackbone=Llama-1 30B, Shots=128-shots2024.050.912—
RLHFBackbone=ChatGLM3 6B2024.050.911—
SFTBackbone=LLaMA2 7B2024.050.91—
SFTModel=LLaMA2, Model Size=7B2024.050.91—
SFTBackbone=Baichuan2 7B2024.050.908—
SFT w. rat.Backbone=Baichuan2 7B2024.050.907—
RLLRBackbone=Bloom 7B2024.050.901—
RLLRModel=Bloom, Model Size=7B2024.050.901—
ICLModel=Llama-2 13b, ICL Setting=1-shot2024.050.901—
RLLRModel=Bloom, Model Size=3B2024.050.898—
RLHFBackbone=Bloom 7B2024.050.897—
RLHFModel=Bloom, Model Size=7B2024.050.897—
SFT w. rat.Backbone=Bloom 7B2024.050.895—
RLLR_MIXEDBackbone=Bloom 7B2024.050.895—
SFT w. rat.Model=Bloom, Model Size=7B2024.050.895—
RLLRMIXEDModel=Bloom, Model Size=7B2024.050.895—
RLLRMIXEDModel=Bloom, Model Size=3B2024.050.894—
ICLModel=Llama-2 7b, ICL Setting=1-shot2024.050.8937—
SFTBackbone=Bloom 7B2024.050.892—
SFTModel=Bloom, Model Size=7B2024.050.892—
SFTBackbone=ChatGLM3 6B2024.050.89—
SFT w. rat.Model=Bloom, Model Size=3B2024.050.887—
RLHFModel=Bloom, Model Size=3B2024.050.886—
SFTModel=Bloom, Model Size=3B2024.050.884—
BERT-12Dim=768, Fine-tuning status=true2021.060.8740.874
kNN-promptNumber of training shots (m)=128, LLM scale=1.5B2024.050.873—
kNN-promptBackbone=GPT-2 1.5B, Shots=128-shots2024.050.873—
FADS-ICLNumber of training shots (m)=64, LLM scale=1.5B2024.050.87—
FADS-ICLcandidate pool size (m)=642024.050.87—
kNN-promptingNumber of training shots (m)=32, LLM scale=1.5B2024.050.864—
kNN-promptingNumber of training shots (m)=128, LLM scale=1.5B2024.050.862—
GPT Large FTBackbone=GPT Large, Shots=N/A (Fine-tuned)2024.050.862—
kNN-promptingBackbone=GPT-2 1.5B, Shots=128-shots2024.050.862—
FADS-ICLcandidate pool size (m)=2562024.050.862—
FADS-ICLNumber of training shots (m)=128, LLM scale=1.5B2024.050.861—
FADS-ICLBackbone=GPT-2 1.5B, Shots=128-shots2024.050.861—
FADS-ICLcandidate pool size (m)=1282024.050.861—
FADS-ICLcandidate pool size (m)=322024.050.859—
kNN-promptNumber of training shots (m)=64, LLM scale=1.5B2024.050.852—
kNN-promptBackbone=GPT-2 0.8B, Shots=128-shots2024.050.85—
FADS-ICLBackbone=GPT-2 0.8B, Shots=128-shots2024.050.85—
kNN-promptingNumber of training shots (m)=8, LLM scale=1.5B2024.050.845—
kNN-promptNumber of training shots (m)=32, LLM scale=1.5B2024.050.845—
kNN-promptingNumber of training shots (m)=64, LLM scale=1.5B2024.050.843—
kNN-promptingNumber of training shots (m)=16, LLM scale=1.5B2024.050.836—
BERT Large FTBackbone=BERT Large, Shots=N/A (Fine-tuned)2024.050.83—
kNN-promptingNumber of training shots (m)=4, LLM scale=1.5B2024.050.829—
FADS-ICLNumber of training shots (m)=32, LLM scale=1.5B2024.050.828—