Update README.md

README.md

@@ -140,13 +140,12 @@ with few-shot examples, each sample will be divided into an independent block. T
 If you find `Block-Attention` useful for your research, please cite our paper:
 
 ```bibtex
-@misc{sun2024blockattentionefficientrag,
-      title={Block-Attention for Efficient RAG}, 
-      author={East Sun and Yan Wang and Lan Tian},
-      year={2024},
-      eprint={2409.15355},
-      archivePrefix={arXiv},
-      primaryClass={cs.LG},
-      url={https://arxiv.org/abs/2409.15355}, 
+@inproceedings{
+    ma2025blockattention,
+    title={Block-Attention for Efficient Prefilling},
+    author={Dongyang Ma and Yan Wang and Tian Lan},
+    booktitle={The Thirteenth International Conference on Learning Representations},
+    year={2025},
+    url={https://openreview.net/forum?id=7zNYY1E2fq}
 }
 ```