Update README.md

README.md

@@ -10,4 +10,4 @@ language:
 To evaluate various pre-training data detection methods in a more challenging scenario, we introduce ArxivMIA, a new benchmark comprising abstracts from the fields of Computer Science (CS) and Mathematics (Math) sourced from Arxiv.
 
 - **Repository:** https://github.com/zhliu0106/probing-lm-data
-- **Paper [optional]:** Probing Language Models for Pre-training Data Detection
+- **Paper:** Probing Language Models for Pre-training Data Detection