Emotion Classification on EmoC
86.8AccuracyLORA
Evaluation results
| Method | Links | |
|---|---|---|
| LORABackbone=Llama2, k (training examples per class)=200, #Param=4.2M2024.02 | 86.8 | |
| AdapterBackbone=Llama2, k (training examples per class)=200, #Param=198M2024.02 | 85.05 | |
| AdapterBackbone=Llama2, k (training examples per class)=5, #Param=198M2024.02 | 84.05 | |
| GNNAVI-GCNBackbone=Llama2, k (training examples per class)=200, #Param=16.8M2024.02 | 82.85 | |
| GNNAVI-SAGEBackbone=GPT2-XL, k (training examples per class)=200, #Param=5.1M2024.02 | 82.68 | |
| GNNAVI-SAGEBackbone=Llama2, k (training examples per class)=200, #Param=33.6M2024.02 | 81.94 | |
| PrefixBackbone=Llama2, k (training examples per class)=200, #Param=39.3M2024.02 | 81.72 | |
| LORABackbone=GPT2-XL, k (training examples per class)=200, #Param=2.5M2024.02 | 80.8 | |
| AdapterBackbone=GPT2-XL, k (training examples per class)=200, #Param=15.4M2024.02 | 80.7 | |
| PrefixBackbone=GPT2-XL, k (training examples per class)=200, #Param=6.1M2024.02 | 80.18 | |
| GNNAVI-SAGEBackbone=Llama2, k (training examples per class)=5, #Param=33.6M2024.02 | 80.12 | |
| FPFTBackbone=Llama2, k (training examples per class)=200, #Param=6.7B2024.02 | 79.9 | |
| GNNAVI-GCNBackbone=GPT2-XL, k (training examples per class)=200, #Param=2.6M2024.02 | 78.82 | |
| GNNAVI-SAGEBackbone=GPT2-XL, k (training examples per class)=5, #Param=5.1M2024.02 | 78.7 | |
| GNNAVI-GCNBackbone=Llama2, k (training examples per class)=5, #Param=16.8M2024.02 | 78.3 | |
| AdapterBackbone=GPT2-XL, k (training examples per class)=5, #Param=15.4M2024.02 | 76 | |
| GNNAVI-GCNBackbone=GPT2-XL, k (training examples per class)=5, #Param=2.6M2024.02 | 75.48 | |
| FPFTBackbone=GPT2-XL, k (training examples per class)=200, #Param=1.6B2024.02 | 73.7 | |
| PrefixBackbone=GPT2-XL, k (training examples per class)=5, #Param=6.1M2024.02 | 73.46 | |
| LORABackbone=Llama2, k (training examples per class)=5, #Param=4.2M2024.02 | 64.2 | |
| FPFTBackbone=Llama2, k (training examples per class)=5, #Param=6.7B2024.02 | 61.92 | |
| FPFTBackbone=GPT2-XL, k (training examples per class)=5, #Param=1.6B2024.02 | 61.3 | |
| PrefixBackbone=Llama2, k (training examples per class)=5, #Param=39.3M2024.02 | 58.56 | |
| Latent Concept LearningBackbone=GPT2-large, Configuration=concept tokens as prefixes2023.01 | 57.3 | |
| LORABackbone=GPT2-XL, k (training examples per class)=5, #Param=2.5M2024.02 | 50.6 | |
| Latent Concept LearningLLM=GPT3-c (6.7B)2023.01 | 50.3 | |
| Latent Concept LearningLLM=GPT2-l (774M)2023.01 | 48.4 | |
| SimilarLLM=GPT3-c (6.7B)2023.01 | 47.4 | |
| UniformLLM=GPT3-c (6.7B)2023.01 | 44.3 | |
| Latent Concept LearningLLM=GPT2-xl (1.5B)2023.01 | 43.6 | |
| Latent Concept LearningLLM=GPT3-b (1.3B)2023.01 | 43.6 | |
| Latent Concept LearningLLM=GPT3-a (350M)2023.01 | 43 | |
| SimilarLLM=GPT3-b (1.3B)2023.01 | 42.9 | |
| SimilarLLM=GPT3-a (350M)2023.01 | 42.4 | |
| UniformLLM=GPT3-b (1.3B)2023.01 | 42 | |
| Latent Concept LearningLLM=GPT3-d (175B)2023.01 | 41.3 | |
| Latent Concept LearningLLM=GPT2-m (355M)2023.01 | 40.4 | |
| SimilarLLM=GPT2-l (774M)2023.01 | 39.9 | |
| SimilarLLM=GPT2-xl (1.5B)2023.01 | 39.9 | |
| UniformLLM=GPT3-d (175B)2023.01 | 39.4 | |
| UniformLLM=GPT2-l (774M)2023.01 | 38.7 | |
| UniformLLM=GPT3-a (350M)2023.01 | 38.6 | |
| UniformLLM=GPT2-xl (1.5B)2023.01 | 38.2 | |
| SimilarLLM=GPT3-d (175B)2023.01 | 37.6 | |
| Latent Concept LearningLLM=GPT2 (124M)2023.01 | 37.2 | |
| SimilarLLM=GPT2-m (355M)2023.01 | 37.2 | |
| SimilarLLM=GPT2 (124M)2023.01 | 36.4 | |
| UniformLLM=GPT2 (124M)2023.01 | 35.3 | |
| UniformLLM=GPT2-m (355M)2023.01 | 34.2 | |
| ICLBackbone=Llama2, k (training examples per class)=52024.02 | 20.18 | |
| ICLBackbone=Llama2, k (training examples per class)=02024.02 | 9.6 | |
| ICLBackbone=GPT2-XL, k (training examples per class)=02024.02 | 6.48 | |
| ICLBackbone=GPT2-XL, k (training examples per class)=52024.02 | 6.3 |