Plain Summarization on GovReport
46.93ROUGE-1Llama-3.1-8B-Instruct
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Llama-3.1-8B-InstructRevision Stage=Initial Summary2026.04 | 46.93 | 86.3 | 19.95 | 13.43 | 16.6 | 55.37 | 43.63 | |
| Qwen3-8BRevision Stage=Initial Summary2026.04 | 46.19 | 85.59 | 19.81 | 14.24 | 19.38 | 56.04 | 38.59 | |
| Llama-3.1-8B-InstructRevision Stage=Round 1, Revision Method=NRLB2026.04 | 45.33 | 85.84 | 17.51 | 12.51 | 15.09 | 59.2 | 45.7 | |
| Llama-3.1-8B-InstructRevision Stage=Round 2, Revision Method=NRLB2026.04 | 44.19 | 85.48 | 15.98 | 11.96 | 14.23 | 60.6 | 47.76 | |
| Qwen3-8BRevision Stage=Round 1, Revision Method=NRLB2026.04 | 41.41 | 85.04 | 15.51 | 13.31 | 17.06 | 63.37 | 45.38 | |
| Qwen3-8BRevision Stage=Round 2, Revision Method=NRLB2026.04 | 40.04 | 84.74 | 13.72 | 12.9 | 16.01 | 64.61 | 49 | |
| GPT-4oRevision Stage=Initial Summary2026.04 | 37.93 | 86.14 | 17.79 | 13.33 | 17.77 | 60.28 | 42.36 | |
| GPT-4oRevision Stage=Round 1, Revision Method=NRLB2026.04 | 36.35 | 85.89 | 13.92 | 12.49 | 15.71 | 66.32 | 53.14 | |
| GPT-4oRevision Stage=Round 2, Revision Method=NRLB2026.04 | 34.73 | 85.63 | 11.58 | 11.96 | 14.2 | 70 | 61.27 |