Randomized Reductions Discovery on RSR-Bench
793RSR CountA-BITWEEN
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| A-BITWEENBackend=Claude-Opus-4.12024.12 | 793 | 1,628 | 26 | 80 | 221 | 378 | 900 | |
| A-BITWEENBackend=Claude-Sonnet-42024.12 | 293 | 729 | 14 | 66 | 94 | 160 | 274 | |
| N-RESEARCHBackend=Claude-Opus-4.12024.12 | 250 | 539 | 172 | 64 | 197 | 286 | 528 | |
| N-RESEARCHBackend=Claude-Sonnet-42024.12 | 191 | 421 | 153 | 60 | 56 | 110 | 203 | |
| N-RESEARCHBackend=GPT-OSS-120B2024.12 | 170 | 360 | 112 | 62 | 5 | 10 | 25 | |
| A-BITWEENBackend=GPT-OSS-120B2024.12 | 157 | 407 | 48 | 59 | 6 | 30 | 252 | |
| V-BITWEENBackend=LR2024.12 | 87 | 87 | 46 | 54 | 0 | 5 | 19 | |
| V-BITWEENBackend=MILP2024.12 | 74 | 74 | 29 | 51 | 0 | 11 | 47 | |
| V-BITWEENBackend=PySR2024.12 | 61 | 61 | 60 | 38 | 113 | 335 | 956 | |
| V-BITWEENBackend=GPLearn2024.12 | 48 | 48 | 54 | 32 | 0 | 140 | 629 |