Summarization on eBay Email
2.062Mean Auto-Rating (r)Llama 3.1 8B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Llama 3.1 8BVersion=r22025.11 | 2.062 | 0.477 | |
| Llama 3.1 8BVersion=r12025.11 | 2.06 | 0.475 | |
| Llama 3.2 3BVersion=r12025.11 | 1.91 | 0.594 | |
| Phi3.5mini 4BVersion=r22025.11 | 1.792 | 0.324 | |
| Llama 3.2 3BVersion=r22025.11 | 1.721 | 0.405 | |
| Llama 3.2 1BVersion=r22025.11 | 1.689 | 0.353 | |
| Phi3.5mini 4BVersion=r12025.11 | 1.675 | 0.207 | |
| Qwen3 4BVersion=r12025.11 | 1.633 | 0.133 | |
| Llama 3.2 1BVersion=r12025.11 | 1.6 | 0.264 | |
| Qwen3 4BVersion=out-of-box2025.11 | 1.594 | 0.094 | |
| Llama 3.1 8BVersion=org2025.11 | 1.585 | — | |
| Qwen3 4BVersion=r22025.11 | 1.535 | 0.035 | |
| Phi3.5mini 4BVersion=out-of-box2025.11 | 1.535 | 0.067 | |
| Qwen3 4BVersion=org2025.11 | 1.5 | — | |
| GPT3.5Version=out-of-box2025.11 | 1.485 | — | |
| Phi3.5mini 4BVersion=org2025.11 | 1.468 | — | |
| Llama 3.2 1BVersion=org2025.11 | 1.336 | — | |
| Llama 3.2 3BVersion=org2025.11 | 1.316 | — | |
| Llama 3.1 8BVersion=out-of-box2025.11 | 1.185 | -0.4 | |
| Llama 3.2 3BVersion=out-of-box2025.11 | 1.182 | -0.134 | |
| Gemma3 1BVersion=r22025.11 | 1.13 | 0.092 | |
| Gemma3 270MVersion=r12025.11 | 1.049 | 0.027 | |
| Gemma3 1BVersion=org2025.11 | 1.038 | — | |
| Gemma3 1BVersion=r12025.11 | 1.027 | -0.011 | |
| Gemma3 1BVersion=out-of-box2025.11 | 1.022 | -0.016 | |
| Gemma3 270MVersion=org2025.11 | 1.022 | — | |
| Gemma3 270MVersion=r22025.11 | 1.022 | 0 | |
| Gemma3 270MVersion=out-of-box2025.11 | 1.014 | -0.008 | |
| Llama 3.2 1BVersion=out-of-box2025.11 | 1 | -0.721 |