Update README.md

README.md

@@ -393,6 +393,7 @@ To speed up Long-context LLMs' inference, approximate and dynamic sparse calcula
 22. [LLMs应用构建一年之心得](https://iangyan.github.io/2024/09/08/building-with-llms-part-1/)
 23. [LLM训练-pretrain](https://zhuanlan.zhihu.com/p/718354385)
 24. [pytorch-llama](https://github.com/hkproj/pytorch-llama): LLaMA 2 implemented from scratch in PyTorch.
+25. [Preference Optimization for Vision Language Models with TRL](https://huggingface.co/blog/dpo_vlm) 【[support model](https://huggingface.co/docs/transformers/model_doc/auto#transformers.AutoModelForVision2Seq)】
 
 <div align="right">
     <b><a href="#Contents">↥ back to top</a></b>