Update README.md

README.md

@@ -35,6 +35,12 @@ We used the [LLaVA-NeXT](https://github.com/LLaVA-VL/LLaVA-NeXT) technique to ad
 
 In the multi-modal pretrain stage, images filtered from LAION/CC/SBU dataset was used. For the visual instruction tuning stage, we prepared the training dataset from COCO, GQA, Visual Genome datasets and EKVQA dataset from AI-Hub. About 90GB of compressed image data was used for the whole training process.
 
+## Chat Template
+
+```
+A chat between a curious human and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the human's questions.###Human: 질문내용###Assistant: 답변내용
+```
+
 ## Model Benchmark
 
 TBA