Update README.md

README.md

@@ -20,6 +20,19 @@ The LLaVE models are 2B parameter multimodal embedding models based on the Aquil
 
 The model have the ability to embed with texts, images, multi-image and videos. 
 
+## MMEB Leaderboard
+We achieved the top ranking on the MMEB leaderboard using only a small amount of data.
+
+![MMEB Leaderboard](./figures/leaderboard.png)
+
+
+## Model Performance
+LLaVE-7B achieved the SOTA performance on MMEB using only 662K training pairs.
+![MMEB](./figures/results.png)
+
+Although LLaVE is trained on image-text data, it can generalize to text-video retrieval tasks in a zero-shot manner and achieve strong performance, demonstrating its remarkable potential for transfer to other embedding tasks.
+<img src="./figures/zero-shot-vr.png" alt="video-retrieve" width="400" height="auto">
+
 ### Quick Start
 
 First clone our github