Mathematical Reasoning on GK EN 2023 (test)
77.9Pass@1Transformers
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TransformersBackbone=R1-Qwen2026.01 | 77.9 | 593.7 | |
| DYNTSBackbone=R1-Qwen2026.01 | 76.4 | 788.5 | |
| R-KVBackbone=R1-Qwen2026.01 | 75.8 | 792.8 | |
| SnapKVBackbone=R1-Qwen2026.01 | 74.9 | 787.4 | |
| H2OBackbone=R1-Qwen2026.01 | 74.1 | 780.1 | |
| TransformersBackbone=R1-Llama2026.01 | 73.1 | 390.9 | |
| DYNTSBackbone=R1-Llama2026.01 | 72.3 | 631.8 | |
| SepLLMBackbone=R1-Qwen2026.01 | 72 | 820.1 | |
| WindowBackbone=R1-Qwen2026.01 | 71.8 | 815.2 | |
| R-KVBackbone=R1-Llama2026.01 | 71.4 | 622.6 | |
| StreamingLLMBackbone=R1-Qwen2026.01 | 71.2 | 824 | |
| SnapKVBackbone=R1-Llama2026.01 | 68.7 | 615.7 | |
| H2OBackbone=R1-Llama2026.01 | 67.5 | 601.8 | |
| SepLLMBackbone=R1-Llama2026.01 | 61.4 | 635 | |
| StreamingLLMBackbone=R1-Llama2026.01 | 53.4 | 624.6 | |
| WindowBackbone=R1-Llama2026.01 | 47 | 622.6 |