Update README.md

README.md

@@ -68,8 +68,6 @@ for various magnet configurations.
 
 English
 
-## Dataset Structure
-
 ### Data Instances
 
 Each instance contains:
@@ -83,8 +81,6 @@ Each instance contains:
 
 The dataset has only a train split with 71,225 examples.
 
-## Dataset Creation
-
 ### Curation Rationale
 
 This dataset was created to provide high-quality simulated magnetic field data for machine learning research in physics and engineering.
@@ -101,8 +97,6 @@ No annotations, only raw simulation data.
 
 None.
 
-## Considerations for Using the Data
-
 ### Social Impact of Dataset
 
 This dataset enables research in electromagnetic field prediction and physics-informed machine learning.
@@ -115,8 +109,6 @@ The data is synthetic and evenly distributed across magnet types.
 
 Data is simulated and may not match real-world measurements exactly.
 
-## Additional Information
-
 ### Dataset Curators
 
 webXOS Team