Update README.md

README.md

@@ -45,7 +45,7 @@ size_categories:
 
 ### Dataset Summary
 
-MemoryCode is a synthetic multi-session dataset that evaluates Large Language Models’ ability to track and execute simple coding instructions amid irrelevant information, simulating real-world collaboration.
+MemoryCode is a synthetic multi-session dataset that evaluates Large Language Models’ ability to track and execute simple coding instructions, simulating real-world collaboration.
  
 - **Paper:** [From Tools to Teammates: Evaluating LLMs in Multi-Session Coding Interactions](https://arxiv.org/abs/2502.13791)
 - **GitHub repository:** [https://github.com/for-ai/MemoryCode](https://github.com/for-ai/MemoryCode)