Sentiment Classification on SST-2 (Delta Metrics)
0.05Delta AccuracyUAT
Evaluation results
| Method | Links | ||
|---|---|---|---|
| UATTarget Model=Qwen2-1.5B, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | 0.05 | 1.33 | |
| AutoPromptTarget Model=Qwen2-1.5B, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | 0.01 | 0.05 | |
| UATTarget Model=Qwen2-1.5B, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | 0 | 1.78 | |
| AutoPromptTarget Model=Qwen2-1.5B, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.01 | 1.1 | |
| UATTarget Model=TinyLlama, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.01 | 0.25 | |
| AutoPromptTarget Model=TinyLlama, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.01 | 0.26 | |
| Calibrated Soft Suffix LearningTarget Model=TinyLlama, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.02 | 0.45 | |
| soft promptTarget Model=Qwen2-1.5B, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.03 | 2.74 | |
| UATTarget Model=Phi-1.5, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.03 | 0.4 | |
| soft promptTarget Model=TinyLlama, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.03 | 0.12 | |
| soft promptTarget Model=Qwen2-1.5B, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.04 | 2.1 | |
| soft promptTarget Model=Phi-1.5, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.05 | 0.01 | |
| UATTarget Model=Phi-1.5, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.06 | 0.45 | |
| AutoPromptTarget Model=TinyLlama, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.06 | 2 | |
| Calibrated Soft Suffix LearningTarget Model=TinyLlama, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.07 | -0.92 | |
| soft promptTarget Model=Phi-1.5, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.1 | 0.7 | |
| AutoPromptTarget Model=Phi-1.5, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.12 | 0.04 | |
| AutoPromptTarget Model=Phi-1.5, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.12 | -0.87 | |
| soft promptTarget Model=TinyLlama, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.13 | 1.58 | |
| Calibrated Soft Suffix LearningTarget Model=Qwen2-1.5B, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.17 | 0.43 | |
| UATTarget Model=TinyLlama, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.22 | 0.72 | |
| Calibrated Soft Suffix LearningTarget Model=Phi-1.5, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.24 | -0.24 | |
| Calibrated Soft Suffix LearningTarget Model=Phi-1.5, Shot Setting=0-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.29 | 0.14 | |
| Calibrated Soft Suffix LearningTarget Model=Qwen2-1.5B, Shot Setting=4-shot, Token Length (K)=4, Seen Model=Qwen2-1.5B-Instruct2025.12 | -0.3 | -0.48 |