Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -136,14 +136,14 @@ Instruction Following Rate (IFR) measures the ability of the model to parse natu
|
|
| 136 |
|
| 137 |
The metric is defined as:
|
| 138 |
|
| 139 |
-
|
| 140 |
IFR = \left( \frac{N_{correct}}{N_{total}} \right) \times 100\%
|
| 141 |
-
|
| 142 |
|
| 143 |
where:
|
| 144 |
|
| 145 |
-
-
|
| 146 |
-
-
|
| 147 |
|
| 148 |
Compared with mainstream open-source models, **I-OSUM-Pangu achieves significantly better performance**:
|
| 149 |
|
|
|
|
| 136 |
|
| 137 |
The metric is defined as:
|
| 138 |
|
| 139 |
+
$$
|
| 140 |
IFR = \left( \frac{N_{correct}}{N_{total}} \right) \times 100\%
|
| 141 |
+
$$
|
| 142 |
|
| 143 |
where:
|
| 144 |
|
| 145 |
+
- $N_{correct}$ represents the number of correctly executed instructions
|
| 146 |
+
- $N_{total}$ represents the total number of evaluation samples
|
| 147 |
|
| 148 |
Compared with mainstream open-source models, **I-OSUM-Pangu achieves significantly better performance**:
|
| 149 |
|