Logic Reasoning on Enigmata
49.2Reasoning ScoreRL_RACES
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RL_RACESBackbone=Qwen3-14B2026.06 | 49.2 | 61.1 | |
| RL_individualBackbone=Qwen3-14B2026.06 | 48.2 | 60.1 | |
| BaseBackbone=Qwen3-14B2026.06 | 47.4 | 58.8 | |
| RL_RACESBackbone=DeepSeek-R1-Distill-Qwen-14B2026.06 | 35.4 | 51.3 | |
| RL_individualBackbone=DeepSeek-R1-Distill-Qwen-14B2026.06 | 34.2 | 48.8 | |
| BaseBackbone=DeepSeek-R1-Distill-Qwen-14B2026.06 | 32.3 | 48.2 |