Creative Writing on Creative Writing Evaluation Set 1.0 (test)
8.6CreativityPhi-3-Mini + Top-p
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 2, Sampling=Top-p2025.09 | 8.6 | 8.8 | 5.5 | 7 | 8.7 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 2, Sampling=Top-H2025.09 | 8.6 | 8.2 | 8.4 | 8.3 | 8.3 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 1, Sampling=Top-H2025.09 | 8.5 | 8.3 | 8.8 | 8.4 | 9.2 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 1, Sampling=Top-H2025.09 | 8.4 | 8.4 | 8 | 8.1 | 8.6 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 2, Sampling=Top-H2025.09 | 8.2 | 8 | 7.4 | 7.6 | 7.8 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 3, Sampling=Top-H2025.09 | 8.2 | 8 | 8 | 8 | 8.3 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 1, Sampling=Top-H2025.09 | 8.2 | 7.8 | 7.2 | 7.6 | 7.8 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 2, Sampling=Top-H2025.09 | 8.2 | 8.2 | 8 | 8 | 8.6 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 3, Sampling=Top-H2025.09 | 8.1 | 9 | 8 | 8 | 9 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 1, Sampling=Top-p2025.09 | 8 | 7.5 | 9 | 7.6 | 8.4 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 1, Sampling=Min-p2025.09 | 7.9 | 7.8 | 8.4 | 7.5 | 8.4 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 3, Sampling=Top-H2025.09 | 7.9 | 7.6 | 8 | 7.3 | 8.4 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 1, Sampling=Top-H2025.09 | 7.9 | 8 | 8.1 | 7.3 | 7.8 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 1, Sampling=Min-p2025.09 | 7.6 | 7.8 | 8.6 | 8.6 | 7.8 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 1, Sampling=Top-H2025.09 | 7.6 | 7.8 | 8.6 | 8 | 8.6 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 2, Sampling=Top-H2025.09 | 7.6 | 7.6 | 7.2 | 7.6 | 7.8 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 3, Sampling=Top-H2025.09 | 7.6 | 7.5 | 7.6 | 7.2 | 8.1 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 2, Sampling=Top-p2025.09 | 7.6 | 8 | 4.6 | 5.8 | 7 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 2, Sampling=Top-H2025.09 | 7.5 | 7.8 | 8.1 | 7 | 7.4 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 1, Sampling=Top-p2025.09 | 7.4 | 7.4 | 8 | 8 | 7.2 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 2, Sampling=Top-H2025.09 | 7.4 | 7.4 | 8 | 7.8 | 8.4 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 2, Sampling=Min-p2025.09 | 7.4 | 7.4 | 8.2 | 7.8 | 8 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 3, Sampling=Top-p2025.09 | 7.4 | 7.1 | 8.4 | 7.1 | 7.8 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 2, Sampling=Min-p2025.09 | 7.4 | 7.5 | 8.8 | 7.8 | 7.8 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 3, Sampling=Top-p2025.09 | 7.4 | 7.6 | 5 | 5.8 | 7.4 | |
| Phi-3-Mini + Top-HLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 3, Sampling=Top-H2025.09 | 7.4 | 7.8 | 8.4 | 7.6 | 8.2 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 2, Sampling=Top-p2025.09 | 7.4 | 6.8 | 8 | 6.8 | 7.1 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 1, Sampling=Top-H2025.09 | 7.4 | 7.4 | 8 | 6.8 | 7.2 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 3, Sampling=Top-p2025.09 | 7.4 | 7.1 | 4.3 | 5.5 | 6.3 | |
| Qwen2.5 + Top-HLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 3, Sampling=Top-H2025.09 | 7.3 | 7.4 | 8.4 | 8.2 | 8.3 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 2, Sampling=Top-p2025.09 | 7.2 | 7 | 8.4 | 6.6 | 7.4 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 2, Sampling=Top-p2025.09 | 7.2 | 6.8 | 7.2 | 6.6 | 7.6 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 1, Sampling=Top-p2025.09 | 7.2 | 7 | 5.6 | 6 | 7 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 2, Sampling=Top-p2025.09 | 7.2 | 6.8 | 7.6 | 6.8 | 7 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 3, Sampling=Top-p2025.09 | 7.2 | 6.9 | 7.3 | 6.2 | 6.7 | |
| Phi-3-Mini + Top-pLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 3, Sampling=Top-p2025.09 | 7 | 6.4 | 8 | 7.6 | 8 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=1.5, Prompt=Prompt 3, Sampling=Min-p2025.09 | 6.9 | 6.6 | 7.7 | 7.2 | 7.2 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 2, Sampling=Min-p2025.09 | 6.8 | 7 | 7.6 | 7.2 | 6.4 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 2, Sampling=Min-p2025.09 | 6.8 | 6.2 | 7.4 | 6.8 | 6.8 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 1, Sampling=Min-p2025.09 | 6.8 | 6.6 | 7 | 6 | 6.8 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 3, Sampling=Min-p2025.09 | 6.7 | 6.5 | 7.2 | 7.1 | 7 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 1, Sampling=Min-p2025.09 | 6.6 | 6.5 | 7.8 | 6.7 | 7.2 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=2.0, Prompt=Prompt 3, Sampling=Min-p2025.09 | 6.6 | 6.6 | 8 | 7 | 7.4 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 3, Sampling=Min-p2025.09 | 6.6 | 6.5 | 5.5 | 6.4 | 6.9 | |
| Phi-3-Mini + Min-pLLM=Phi-3-Mini, Temperature=1.0, Prompt=Prompt 3, Sampling=Min-p2025.09 | 6.4 | 5.5 | 8 | 6.2 | 7.2 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 2, Sampling=Min-p2025.09 | 6.4 | 6.4 | 7 | 6.2 | 6.4 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 3, Sampling=Min-p2025.09 | 6.4 | 6 | 7 | 5.8 | 6.6 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 2, Sampling=Min-p2025.09 | 6.2 | 6.3 | 7.8 | 5.8 | 5.9 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 1, Sampling=Min-p2025.09 | 6.1 | 5.4 | 6 | 5.1 | 4.9 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 3, Sampling=Top-p2025.09 | 6 | 5.3 | 8.4 | 6.4 | 6.8 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=1.5, Prompt=Prompt 1, Sampling=Top-p2025.09 | 6 | 5.3 | 7.8 | 5.6 | 5.6 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=2.0, Prompt=Prompt 1, Sampling=Top-p2025.09 | 6 | 5.4 | 3.8 | 3.6 | 4.4 | |
| Qwen2.5 + Top-pLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 1, Sampling=Top-p2025.09 | 5 | 4.8 | 7.2 | 5.4 | 5.2 | |
| Qwen2.5 + Min-pLLM=Qwen2.5, Temperature=1.0, Prompt=Prompt 1, Sampling=Min-p2025.09 | 4.8 | 4 | 6.2 | 4.2 | 4.2 |