Counseling Accuracy Evaluation on CounselingBench
63.1AccuracyLlama3-SQP
Evaluation Results
| Method | Links | |
|---|---|---|
| Llama3-SQPPrompting strategy=Few-Shot (FS), Model state=Fine-tuned2026.04 | 63.1 | |
| Llama3-G2CPrompting strategy=Few-Shot (FS), Model state=Fine-tuned2026.04 | 62.5 | |
| Llama3-MAGPrompting strategy=Zero-Shot (ZS), Model state=Fine-tuned2026.04 | 61.6 | |
| Llama3-MAGPrompting strategy=Few-Shot (FS), Model state=Fine-tuned2026.04 | 61.6 | |
| Llama3-G2CPrompting strategy=Zero-Shot (ZS), Model state=Fine-tuned2026.04 | 61.6 | |
| Llama3-SQPPrompting strategy=Zero-Shot (ZS), Model state=Fine-tuned2026.04 | 61.5 | |
| CAMELPrompting strategy=Few-Shot (FS), Model state=Fine-tuned2026.04 | 61.4 | |
| CAMELPrompting strategy=Zero-Shot (ZS), Model state=Fine-tuned2026.04 | 61.3 | |
| Llama3-MAGPrompting strategy=Few-Shot Chain-of-Thought (FS-CoT), Model state=Fine-tuned2026.04 | 58.2 | |
| Llama3-SQPPrompting strategy=Few-Shot Chain-of-Thought (FS-CoT), Model state=Fine-tuned2026.04 | 56.1 | |
| Llama3-G2CPrompting strategy=Few-Shot Chain-of-Thought (FS-CoT), Model state=Fine-tuned2026.04 | 56 | |
| CAMELPrompting strategy=Few-Shot Chain-of-Thought (FS-CoT), Model state=Fine-tuned2026.04 | 47 |