Update README.md

README.md

@@ -91,4 +91,17 @@ cont = model.generate(
 )
 text_outputs = tokenizer.batch_decode(cont, skip_special_tokens=True)
 print(text_outputs[0])
+```
+
+## Citation
+
+If you find our repo useful for your research, please consider citing our paper:
+
+```bibtex
+@article{sun2025llava,
+  title={LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMs},
+  author={Sun, Boyuan and Zhao, Jiaxing and Wei, Xihan and Hou, Qibin},
+  journal={arXiv preprint arXiv:2506.21862},
+  year={2025}
+}
 ```