Update README.md

README.md

@@ -16,6 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama-3-8b-dpo-ultrafeedback-decrease_linear-1.0to0.95
 
+This is a model released from the preprint: [DPO-Shift: Shifting the Distribution of Direct Preference Optimization](https://arxiv.org/abs/2502.07599). Please refer to our [repository](https://github.com/Meaquadddd/DPO-Shift) for more details.
+
+
 This model is a fine-tuned version of [princeton-nlp/Llama-3-Base-8B-SFT](https://huggingface.co/princeton-nlp/Llama-3-Base-8B-SFT) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.5619