Add model card and metadata

README.md

@@ -1,14 +1,28 @@
 ---
-license: apache-2.0
 base_model:
 - mistralai/Mistral-7B-Instruct-v0.2
+license: apache-2.0
+library_name: transformers
+pipeline_tag: text-generation
 ---
 
+# RESM-Mistral-7B
+
+This model is a 3H-aligned (Helpful, Honest, and Harmless) version of Mistral-7B-Instruct-v0.2, developed as part of the research presented in the paper [Mix Data or Merge Models? Balancing the Helpfulness, Honesty, and Harmlessness of Large Language Model via Model Merging](https://huggingface.co/papers/2502.06876).
+
+## Model Description
+The model aims to achieve a balanced alignment across three critical dimensions: Helpfulness, Honesty, and Harmlessness (3H optimization). It was created using a novel merging method called **RESM** (**R**eweighting **E**nhanced task **S**ingular **M**erging), which utilizes outlier weighting and sparsity-aware rank selection strategies to address challenges such as preference noise accumulation and layer sparsity adaptation inherent in model merging.
+
+- **Developed by:** Jinluan Yang, Dingnan Jin, Anke Tang, Li Shen, Didi Zhu, Zhengyu Chen, Daixin Wang, Qing Cui, Zhiqiang Zhang, Jun Zhou, Fei Wu, Kun Kuang
+- **Base Model:** [Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2)
+- **Method:** RESM (Reweighting Enhanced task Singular Merging)
+
 ## Citation
-```
+```bibtex
 @article{yang2025mix,
   title={Mix Data or Merge Models? Balancing the Helpfulness, Honesty, and Harmlessness of Large Language Model via Model Merging},
   author={Yang, Jinluan and Jin, Dingnan and Tang, Anke and Shen, Li and Zhu, Didi and Chen, Zhengyu and Wang, Daixin and Cui, Qing and Zhang, Zhiqiang and Zhou, Jun and others},
   journal={arXiv preprint arXiv:2502.06876},
   year={2025}
-}
+}
+```