Revision Generation on ITERATER doc
50.13SARIFull-Finetuning
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Full-FinetuningModel=Qwen2.5-14B2026.01 | 50.13 | 55.25 | 38.6 | 47.99 | |
| Intention-TuningModel=Llama3.1-8B2026.01 | 48.77 | 54.86 | 52.08 | 51.9 | |
| Full-FinetuningModel=Llama3.1-8B2026.01 | 47.61 | 54.79 | 51.76 | 51.39 | |
| IR-BaselineModel=Llama3.1-8B2026.01 | 46.94 | 54.74 | 48.79 | 50.16 | |
| Intention-TuningModel=Qwen2.5-14B2026.01 | 46.14 | 54.53 | 40.67 | 47.11 | |
| Intention-TuningModel=Mistral-7B2026.01 | 45.4 | 54.58 | 35.15 | 45.04 | |
| IR-BaselineModel=Qwen2.5-14B2026.01 | 45.07 | 51.8 | 39.51 | 45.46 | |
| IR-BaselineModel=Mistral-7B2026.01 | 45.03 | 54.81 | 34.21 | 44.68 | |
| Full-FinetuningModel=Mistral-7B2026.01 | 42.85 | 53.71 | 32.86 | 43.14 | |
| BART-BaselineModel=N/A2026.01 | 41.68 | 47.57 | 48.22 | 45.82 | |
| IST-BaselineModel=Llama3.1-8B2026.01 | 41.68 | 54.76 | 48.98 | 48.47 | |
| IST-BaselineModel=Qwen2.5-14B2026.01 | 41.06 | 53.84 | 38.93 | 44.61 | |
| CoT-BaselineModel=Llama3.1-8B2026.01 | 39.24 | 18.45 | 47.66 | 35.12 | |
| ICL-BaselineModel=Llama3.1-8B2026.01 | 39.03 | 29.58 | 54.01 | 40.87 | |
| LISA-BaselineModel=Llama3.1-8B2026.01 | 38.66 | 54.03 | 46.11 | 46.27 | |
| ICL-BaselineModel=Qwen2.5-14B2026.01 | 38.28 | 27.86 | 51.85 | 39.33 | |
| CoT-BaselineModel=Qwen2.5-14B2026.01 | 38.09 | 19 | 52.63 | 36.57 | |
| IST-BaselineModel=Mistral-7B2026.01 | 38.03 | 54.53 | 30.9 | 41.15 | |
| ICL-BaselineModel=Mistral-7B2026.01 | 37.93 | 23.9 | 53.69 | 38.51 | |
| LISA-BaselineModel=Qwen2.5-14B2026.01 | 37.63 | 53.89 | 32.76 | 41.43 | |
| CoT-BaselineModel=Mistral-7B2026.01 | 36.74 | 20.42 | 52.45 | 36.54 | |
| LISA-BaselineModel=Mistral-7B2026.01 | 35.86 | 53.91 | 35.66 | 41.81 | |
| Copy-BaselineModel=N/A2026.01 | 27.17 | 53.49 | 0 | 26.89 |