ML Research Engineering on PaperBench
33.26Adaptive Pruning Score
AiScientist
Updated 4d ago
Evaluation Results
| Method | Links | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2026.04 | 33.26 | 49.47 | 61.11 | 30.02 | 26.46 | 28.98 | 8.34 | 30.1 | 28.53 | 40.55 | 58.76 | 10.18 | 28.66 | 44.13 | 31.69 | 49.32 | 14.81 | 42.1 | 27.33 | 30.82 | 33.73 | 12.2 | |
2026.04 | 30.82 | 33.78 | 51.45 | 23.55 | 9.8 | 21.6 | 3.71 | 20.68 | 22.55 | 32.49 | 22.18 | 6.56 | 22.43 | 36.93 | 6.99 | 27.2 | 3.69 | 32.18 | 17.81 | 25.14 | 22.58 | 4.9 | |
2026.04 | 27.25 | 46.29 | 56.59 | 33.79 | 23.09 | 35.21 | 10.11 | 27.9 | 30.23 | 29.95 | 49.92 | 10.87 | 18.28 | 36.77 | 19.85 | 64.94 | 20.13 | 18.81 | 32.45 | 17.87 | 30.52 | 15.67 | |
2026.04 | 24.53 | 20.86 | 48.46 | 15.43 | 12.59 | 21.67 | 5.87 | 17.75 | 12.97 | 14.86 | 26.63 | 10.43 | 15.45 | 25.39 | 11.45 | 53.51 | 8.37 | 17.04 | 15.27 | 6.61 | 19.26 | 6.25 | |
2026.04 | 11.93 | 44.43 | 47.91 | 19.28 | 12.5 | 16.67 | 6.7 | 22.74 | 26.15 | 34.96 | 25.77 | 0.27 | 27.56 | 41.26 | 4.95 | 35.53 | 8.81 | 28.06 | 21.19 | 10.75 | 22.37 | 54.9 | |
2026.04 | 3.05 | 45.13 | 45.04 | 8.3 | 12.44 | 23.89 | 4.15 | 15.26 | 18.3 | 21.89 | 30.81 | 8.88 | 10.43 | 33.34 | 12.65 | 60.24 | 13.69 | 17.37 | 18.13 | 8.99 | 20.6 | 27.44 |