Update README.md

README.md

@@ -32,6 +32,8 @@ This project presents a fine-tuned model based on **Deepseel-R1-Distill-Qwen-1.5
 
 ## Usage
 The model is currently in development, and detailed usage instructions will be provided in future updates.
+Demo input:
+"<|im_start|>system\nPlease reason step by step, and put your final answer within \\boxed{}.<|im_end|>\n<|im_start|>user\nComplete the visitor's travel trajectory in Japan based on the following description. A visitor from Australia traveled in Japan. We have two clues about this visitor: (1) This visitor was visiting Uruma Shi at 2023-07-15 03:12; (2) This visitor spent about 4 days in Japan.\nTask: Complete the travel trajectory of this visitor after detailed reasoning process. The reasoning processes consists of multiple steps which are helpful for this task. They are enclosed within two kinds of tags <think about travel rationality> </think about travel rationality>; <think about format correctness> </think about format correctness>. Afer the reasoning process, put the final answer  within \\boxed{{}} and the answer should follow the format: Visited {location1} at {time1};...; Visited {location2} at {time2}. End of journey.<|im_end|>\n<|im_start|>assistant"
 
 ## Contribution
 If you're interested in contributing or providing feedback, feel free to reach out or submit issues. Your input is valuable in refining the model for better performance.