Update README.md

README.md

@@ -11,8 +11,7 @@ pinned: false
   <img src="./toksuite-pipeline.png" alt="TokSuite Logo"/>
 </p>
 
-**TokSuite** is a comprehensive benchmark covering real-world perturbations that change tokenization (**left**), and **14 models** that share the same initialization, architecture, and
-training data but differ **only in their tokenizers** (**right**).
+**TokSuite** is a collection of models and a benchmark designed to isolate and study the impact of tokenization on language model behavior. It includes **fourteen models** that share the same architecture, data, training budget, and initialization but differ only in their tokenizers, along with a benchmark that evaluates performance under real-world text perturbations that affect tokenization.
 
 The left panel illustrates how different tokenizers fragment the word *“doctor”* when subjected to OCR errors, orthographic mistakes, semantic equivalents, emoji substitution, and multilingual
 translations. Each colored box represents a single token across the **Gemma-2** (yellow), **GPT-2** (blue), **TokenMonster** (green), and **XGLM** (red) tokenizers.