Multi-domain Evaluation on MMLU, GSM8K, MATH, MedQA, and MedMCQA
56.9Math AccuracyType-based (Math)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Type-based (Math)Model Group=Type-based, Configuration Identifier=(iii) Math2025.10 | 56.9 | 56.1 | 56.5 | |
| Com. 2Model Group=(b) Mathematical2025.10 | 56.7 | 63.7 | 60.2 | |
| Single-modelModel Group=Single-model2025.10 | 56.1 | 67.1 | 61.6 | |
| Com. 1Model Group=(c) Medical2025.10 | 54.8 | 66.3 | 60.5 | |
| Com. 2Model Group=(a) General2025.10 | 52.2 | 69.8 | 61 | |
| All-modelsModel Group=All-models2025.10 | 51.8 | 70.1 | 60.9 | |
| Type-based (General)Model Group=Type-based, Configuration Identifier=(i) General2025.10 | 50.9 | 67.9 | 59.4 | |
| Com. 1Model Group=(a) General2025.10 | 50.5 | 68.5 | 59.5 | |
| Random@3modelsModel Group=Random@3models2025.10 | 48.2 | 60.7 | 54.4 | |
| Com. 2Model Group=(c) Medical2025.10 | 44.5 | 73.8 | 59.1 | |
| Com. 3Model Group=(b) Mathematical2025.10 | 41.3 | 49.8 | 45.5 | |
| Type-based (Med)Model Group=Type-based, Configuration Identifier=(iv) Med2025.10 | 39.3 | 76.4 | 57.9 | |
| Com. 1Model Group=(b) Mathematical2025.10 | 37.4 | 72.2 | 54.8 | |
| Com. 3Model Group=(a) General2025.10 | 36.5 | 37.9 | 37.2 | |
| Com. 3Model Group=(c) Medical2025.10 | 36.5 | 37.9 | 37.2 |