Reasoning on AGIEval English
74.4Score (%)Pair+
Evaluation Results
| Method | Links | |
|---|---|---|
| Pair+Backbone=olmo-3.1-32b-it-sft2026.04 | 74.4 | |
| Soft+Backbone=olmo-3.1-32b-it-sft2026.04 | 74.4 | |
| MPO+Backbone=olmo-3.1-32b-it-sft2026.04 | 74.3 | |
| DPOBackbone=olmo-3.1-32b-it-sft2026.04 | 73.6 | |
| Mag+Backbone=olmo-3.1-32b-it-sft2026.04 | 72.6 | |
| RFTBackbone=olmo-3.1-32b-it-sft2026.04 | 69 | |
| BaseBackbone=olmo-3.1-32b-it-sft2026.04 | 67.7 | |
| Pair+Backbone=olmo-3-7b-it-sft2026.04 | 61.8 | |
| Soft+Backbone=olmo-3-7b-it-sft2026.04 | 61.4 | |
| MPO+Backbone=olmo-3-7b-it-sft2026.04 | 60.9 | |
| Mag+Backbone=olmo-3-7b-it-sft2026.04 | 59.4 | |
| DPOBackbone=olmo-3-7b-it-sft2026.04 | 58.8 | |
| RFTBackbone=olmo-3-7b-it-sft2026.04 | 55.8 | |
| BaseBackbone=olmo-3-7b-it-sft2026.04 | 54.1 | |
| MPO+Backbone=gemma-3-4b-sft2026.04 | 51.2 | |
| Pair+Backbone=gemma-3-4b-sft2026.04 | 51.2 | |
| Soft+Backbone=gemma-3-4b-sft2026.04 | 51.2 | |
| Mag+Backbone=gemma-3-4b-sft2026.04 | 50.5 | |
| DPOBackbone=gemma-3-4b-sft2026.04 | 50.3 | |
| BaseBackbone=gemma-3-4b-sft2026.04 | 49.4 | |
| RFTBackbone=gemma-3-4b-sft2026.04 | 49 |