update arxiv url

Author: nanamma

README.md

@@ -1,19 +1,20 @@
 <div align="center">
 
 <h2>
-    <a href="https://github.com/OpenGVLab/RIVER">[ICLR 2026] RIVER: A Real-Time Interaction Benchmark for Video LLMs</a>
+    <a href="https://arxiv.org/abs/2603.03985">[ICLR 2026] RIVER: A Real-Time Interaction Benchmark for Video LLMs</a>
 </h2>
 
 Yansong Shi, Qingsong Zhao, Tianxiang Jiang, Xiangyu Zeng, Yi Wang, Limin Wang
 
-<img src="assets/RIVER.png" width="80" alt="RIVER logo">
+<img src="assets/RIVER logo.png" width="80" alt="RIVER logo">
 
 </div>
 
 
 ## Introduction
 This project introduces **RIVER Bench**, the first benchmark designed to evaluate the real-time interactive capabilities of Video Large Language Models through streaming video perception, featuring novel tasks for memory, live-perception, and proactive response.
-![RIVER](assets/river.pdf)
+
+![RIVER](assets/river.jpg)
 
 ## Dataset Preparation
 |Dataset       |URL|
@@ -24,3 +25,17 @@ This project introduces **RIVER Bench**, the first benchmark designed to evaluat
 |Ego4D         |https://github.com/facebookresearch/Ego4d|
 |QVHighlights  |https://github.com/jayleicn/moment_detr|
 
+## Citation
+
+If you find this project useful in your research, please consider cite:
+```BibTeX
+@misc{shi2026riverrealtimeinteractionbenchmark,
+      title={RIVER: A Real-Time Interaction Benchmark for Video LLMs}, 
+      author={Yansong Shi and Qingsong Zhao and Tianxiang Jiang and Xiangyu Zeng and Yi Wang and Limin Wang},
+      year={2026},
+      eprint={2603.03985},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2603.03985}, 
+}
+```