Text Generation on FuseEval English Scenario
31.19ROUGE-1SpecEM
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| SpecEMModel Category=Methods of Ensembling Base LLMs2024.12 | 31.19 | 14.4 | 28.86 | 5.81 | -2.88 | 73.34 | 3.98 | |
| Mixtral-8x7B-instructModel Category=Larger LLMs2024.12 | 29.04 | 12.25 | 26.75 | 4.08 | -2.81 | 72.19 | 4.16 | |
| PairRankModel Category=Methods of Ensembling Base LLMs2024.12 | 28.21 | 10.86 | 25.94 | 4.99 | -2.86 | 72.09 | 6.84 | |
| UniTEModel Category=Methods of Ensembling Base LLMs2024.12 | 27.81 | 10.91 | 25.73 | 4.53 | -2.9 | 71.77 | 6 | |
| Mistral-7B-instruct-v0.3Model Category=Base LLMs2024.12 | 27.75 | 10.75 | 25.57 | 4.82 | -2.94 | 71.88 | 7.62 | |
| MOAModel Category=Methods of Ensembling Base LLMs2024.12 | 27.61 | 11.3 | 25.47 | 5.12 | -2.88 | 71.9 | 7.48 | |
| Qwen2-72B-instructModel Category=Larger LLMs2024.12 | 27.26 | 11.23 | 25.11 | 4.29 | -2.76 | 71.73 | 4.21 | |
| MBRModel Category=Methods of Ensembling Base LLMs2024.12 | 27.12 | 10.4 | 25.33 | 4.56 | -2.89 | 71.63 | 7.66 | |
| Llama-3-70B-instructModel Category=Larger LLMs2024.12 | 26.77 | 10.87 | 24.56 | 4.1 | -2.84 | 70.98 | 5.22 | |
| Qwen2-7B-instructModel Category=Base LLMs2024.12 | 26.62 | 10.81 | 24.49 | 3.86 | -2.94 | 71.44 | 8.48 | |
| Glm-4-9B-instructModel Category=Base LLMs2024.12 | 25.85 | 10.26 | 23.9 | 3.48 | -2.96 | 70.51 | 9.24 | |
| Gemma-2-9B-instructModel Category=Base LLMs2024.12 | 25.31 | 10.01 | 23.59 | 4.19 | -2.93 | 71.52 | 9.01 | |
| Llama-3-8B-instructModel Category=Base LLMs2024.12 | 25.16 | 9.77 | 23.31 | 3.57 | -2.98 | 69.99 | 9.52 | |
| GF (Mistral)Model Category=Methods of Ensembling Base LLMs2024.12 | 24.92 | 9.58 | 22.97 | 3.92 | -2.93 | 70.38 | 8.56 | |
| GF (Qwen2)Model Category=Methods of Ensembling Base LLMs2024.12 | 23.08 | 8.92 | 21.28 | 3.19 | -2.95 | 69.7 | 10.1 | |
| GF (Gemma-2)Model Category=Methods of Ensembling Base LLMs2024.12 | 21.81 | 7.66 | 20.08 | 3 | -3.02 | 68.2 | 10.11 |