Update README.md

README.md

@@ -46,6 +46,12 @@ For fine-grained performance evaluation results and detailed discussion, check o
 * 🗞️ [Paper](https://arxiv.org/abs/2412.03205)
 * 👾 [Evaluation Code at GitHub](https://github.com/Toloka/u-math/)
 
+### Key Features
+* **Topics Covered**: Precalculus, Algebra, Differential Calculus, Integral Calculus, Multivariable Calculus, Sequences & Series.
+* **Problem Format**: Free-form answer with LLM judgement
+* **Evaluation Metrics**: Accuracy; splits by subject and text-only vs multimodal problem type.
+* **Curation**: Original problems composed by math professors and used in university curricula, samples validated by math experts at [Toloka AI](https://toloka.ai), [Gradarius](https://www.gradarius.com)
+
 ### Use it
 
 ```python
@@ -53,12 +59,6 @@ from datasets import load_dataset
 ds = load_dataset('toloka/u-math', split='test')
 ```
 
-### Key Features
-* **Topics Covered**: Precalculus, Algebra, Differential Calculus, Integral Calculus, Multivariable Calculus, Sequences & Series.
-* **Problem Format**: Free-form answer with LLM judgement
-* **Evaluation Metrics**: Accuracy; splits by subject and text-only vs multimodal problem type.
-* **Curation**: Original problems composed by math professors and used in university curricula, samples validated by math experts at [Toloka AI](https://toloka.ai), [Gradarius](https://www.gradarius.com)
-
 ### Dataset Fields
 `uuid`: problem id \
 `has_image`: a boolean flag on whether the problem is multimodal or not \