Update README.md

README.md

@@ -166,10 +166,10 @@ Filters and dedup are implemented as described above (length, CR, Jaccard, and h
 Please cite **both** the widely used WikiLarge reference **and** this cleaned subset:
 
 **Original alignment (WikiLarge):**
-Zhang, X., & Lapata, M. (2017). *Sentence Simplification with Deep Reinforcement Learning*. In EMNLP 2017 (pp. 584–594). Association for Computational Linguistics. [doi:10.18653/v1/D17-1062]
+Zhang, X., & Lapata, M. (2017). *Sentence Simplification with Deep Reinforcement Learning*. In EMNLP 2017 (pp. 584–594). Association for Computational Linguistics. doi:10.18653/v1/D17-1062
 
 **This dataset (cleaned subset):**  
-Cohen, E., Bul, I., Inbar, D., & Loewenbach, O. (2026). Simplify-This: A Comparative Analysis of Prompt-Based and Fine-Tuned LLMs. arXiv. [arxiv.org/abs/2601.05794]
+Cohen, E., Bul, I., Inbar, D., & Loewenbach, O. (2026). Simplify-This: A Comparative Analysis of Prompt-Based and Fine-Tuned LLMs. arXiv. arxiv.org/abs/2601.05794
 
 WikiLarge BibTeX: