List your company
ResearchFormerly WizWand

Emotion Recognition in Conversation on MELD Standard (test)

69.15Weighted F1InstructERC
6.718822.926939.13555.3431Sep 21, 2023
Updated 8mo ago

Evaluation results

MethodLinks
InstructERCBackbone=LLaMA2-7B, Evaluation Setting=LoRA Fine-tuned2023.0969.15
InstructERCBackbone=LLaMA-7B, Evaluation Setting=LoRA Fine-tuned2023.0967.62
SkierModel Category=Discriminant Model2023.0967.39
DualGATsModel Category=Discriminant Model2023.0966.9
MPLPModel Category=Discriminant Model2023.0966.51
SACLModel Category=Discriminant Model2023.0966.45
EmotionICModel Category=Discriminant Model2023.0966.4
SPCL+CLModel Category=Discriminant Model2023.0966.35
LoRABackbone=LLaMA-7B, Evaluation Setting=Fine-tuned2023.0966.15
LoRABackbone=LLaMA2-7B, Evaluation Setting=Fine-tuned2023.0965.84
InstructERCBackbone=ChatGLM2-6B, Evaluation Setting=LoRA Fine-tuned2023.0965.58
InstructERCBackbone=ChatGPT-3.5, Evaluation Setting=Zero-shot2023.0965.07
LoRABackbone=ChatGLM2-6B, Evaluation Setting=Fine-tuned2023.0964.85
InstructERCBackbone=ChatGLM-6B, Evaluation Setting=LoRA Fine-tuned2023.0946.41
LoRABackbone=ChatGLM-6B, Evaluation Setting=Fine-tuned2023.0940.54
InstructERCBackbone=ChatGLM-6B, Evaluation Setting=Zero-shot2023.0938.8
InstructERCBackbone=ChatGLM2-6B, Evaluation Setting=Zero-shot2023.0921.8
InstructERCBackbone=LLaMA2-7B, Evaluation Setting=Zero-shot2023.0916.28
InstructERCBackbone=LLaMA-7B, Evaluation Setting=Zero-shot2023.099.12