Explanation Generation on ICEWS14 (test)
40.54BLEU-4GETER
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| GETERbackbone=Llama3-8B-Instruct2025.05 | 40.54 | 52.54 | 53.87 | 84.75 | |
| GETERbackbone=Mistral-7B-Instruct2025.05 | 40.21 | 51.84 | 54.9 | 84.65 | |
| GETERbackbone=Qwen2.5-7B-Instruct2025.05 | 39.78 | 51.46 | 55.03 | 84.53 | |
| Qwen2.5-7B-Instructtraining_protocol=LoRA, reasoning_chains=true2025.05 | 39.59 | 51.48 | 53.3 | 84.35 | |
| Llama3-8B-Instructtraining_protocol=LoRA, reasoning_chains=true2025.05 | 39.21 | 50.96 | 54.03 | 84.28 | |
| Mistral-7B-Instructtraining_protocol=LoRA, reasoning_chains=true2025.05 | 38.81 | 50.81 | 52.62 | 84.02 | |
| Qwen2.5-7B-Instructtraining_protocol=LoRA, reasoning_chains=false2025.05 | 28.17 | 40.22 | 45.2 | 80.12 | |
| Mistral-7B-Instructtraining_protocol=LoRA, reasoning_chains=false2025.05 | 28.01 | 39.84 | 45.7 | 80.34 | |
| Llama3-8B-Instructtraining_protocol=LoRA, reasoning_chains=false2025.05 | 27.73 | 39.71 | 45.94 | 80.16 | |
| GPT-4otraining_protocol=zero-shot, reasoning_chains=true2025.05 | 22.94 | 41.04 | 37.24 | 79.25 | |
| Qwen2.5-7B-Instructtraining_protocol=zero-shot, reasoning_chains=true2025.05 | 11.18 | 28.49 | 27.98 | 72.28 | |
| GPT-4otraining_protocol=zero-shot, reasoning_chains=false2025.05 | 10.78 | 23.82 | 31.14 | 68.16 | |
| Llama3-8B-Instructtraining_protocol=zero-shot, reasoning_chains=true2025.05 | 9.7 | 30.19 | 26.6 | 70.25 | |
| Mistral-7B-Instructtraining_protocol=zero-shot, reasoning_chains=true2025.05 | 9.19 | 28.36 | 25.7 | 71.63 | |
| Qwen2.5-7B-Instructtraining_protocol=zero-shot, reasoning_chains=false2025.05 | 7.43 | 19.73 | 30.82 | 66.03 | |
| Mistral-7B-Instructtraining_protocol=zero-shot, reasoning_chains=false2025.05 | 7.17 | 19.4 | 24.27 | 65.46 | |
| Llama3-8B-Instructtraining_protocol=zero-shot, reasoning_chains=false2025.05 | 4.35 | 16.32 | 16.71 | 61.35 |