Rebuttal Generation on Re2 (test)
5.88Judge ScoreDRPG
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| DRPGBase Model=GPT-oss-20B2026.01 | 5.88 | 75.53 | 77.98 | 57.43 | 51.88 | 56.79 | — | 1,067 | |
| DecompBase Model=GPT-oss-20B2026.01 | 5.8 | — | 72.86 | — | 44.03 | 47.89 | 42.57 | 1,012 | |
| DRPGBase Model=Qwen3-8B2026.01 | 5.78 | 60.65 | 65.53 | 56.63 | 58.09 | 58.21 | — | 1,054 | |
| DecompBase Model=Qwen3-8B2026.01 | 5.75 | — | 59.61 | — | 46.77 | 44.11 | 43.37 | 991 | |
| DRGBase Model=Qwen3-8B2026.01 | 5.75 | — | 62.32 | 53.23 | — | 45.51 | 41.91 | 1,004 | |
| DRGBase Model=GPT-oss-20B2026.01 | 5.75 | — | 75.61 | 55.97 | — | 55.53 | 48.02 | 1,054 | |
| DirectBase Model=GPT-oss-20B2026.01 | 5.73 | 40.28 | — | 27.14 | 24.39 | 21.39 | 22.02 | 837 | |
| REALDescription=Human-written rebuttals2026.01 | 5.72 | — | — | — | — | — | — | — | |
| Jiu-JitsuBase Model=Qwen3-8B2026.01 | 5.72 | — | 57.52 | 55.89 | 54.49 | — | 41.79 | 1,013 | |
| Jiu-JitsuBase Model=GPT-oss-20B2026.01 | 5.72 | — | 78.61 | 52.11 | 44.47 | — | 43.21 | 1,029 | |
| DecompBase Model=Mixtral-8x7B2026.01 | 5.68 | — | 82.29 | — | 31.31 | 28.8 | 27.82 | 959 | |
| DRPGBase Model=Mixtral-8x7B2026.01 | 5.68 | 51.42 | 86.62 | 72.18 | 55.33 | 55.59 | — | 1,119 | |
| DRGBase Model=LLaMa3.3-70B2026.01 | 5.68 | — | 81.73 | 57.22 | — | 50.88 | 45.91 | 1,065 | |
| DRPGBase Model=LLaMa3.3-70B2026.01 | 5.68 | 65.44 | 90.93 | 59.39 | 54.09 | 58.72 | — | 1,109 | |
| DecompBase Model=LLaMa3.3-70B2026.01 | 5.67 | — | 89.9 | — | 42.78 | 46.74 | 40.61 | 1,040 | |
| Jiu-JitsuBase Model=LLaMa3.3-70B2026.01 | 5.67 | — | 88.51 | 53.26 | 49.12 | — | 41.28 | 1,058 | |
| DRGBase Model=Mixtral-8x7B2026.01 | 5.66 | — | 87.94 | 68.69 | — | 49.65 | 44.67 | 1,088 | |
| Jiu-JitsuBase Model=Mixtral-8x7B2026.01 | 5.65 | — | 87.85 | 71.2 | 50.35 | — | 44.41 | 1,093 | |
| DirectBase Model=Qwen3-8B2026.01 | 5.63 | 47.81 | — | 40.59 | 37.68 | 42.48 | 34.47 | 936 | |
| DirectBase Model=Mixtral-8x7B2026.01 | 5.63 | 49.04 | — | 17.71 | 12.06 | 12.15 | 13.18 | 738 | |
| DirectBase Model=LLaMa3.3-70B2026.01 | 5.6 | 50.09 | — | 10.1 | 18.27 | 11.49 | 9.07 | 725 |