Update README.md

README.md

@@ -6,7 +6,7 @@ This model is a 7B [Self-RAG](https://selfrag.github.io/) model that generates o
 
 Self-RAG is trained on our instruction-following corpora with interleaving passages and reflection tokens using the standard next-token prediction objective, enabling efficient and stable learning with fine-grained feedback.  
 At inference, we leverage reflection tokens covering diverse aspects of generations to sample the best output aligning users' preferences. 
-See full descriptions in [our paper](https://akariasai.github.io/files/adaptive_retrieval_augmented_lm_arxiv.pdf). 
+See full descriptions in [our paper](hhttps://arxiv.org/abs/2310.11511). 
 
 ## Usage
 Here, we show an easy way to quickly download our model from HuggingFace and run with `vllm` with pre-given passages. Make sure to install dependencies listed at [self-rag/requirements.txt](https://github.com/AkariAsai/self-rag/requirements.txt). 
@@ -49,12 +49,19 @@ or
 If you have additional input. 
 You can insert paragraphs anywhere after `### Response:\n"`, but make sure to mark paragraphs as paragraph tokens (i.e., `<paragraph>{0}</paragraph>`).
 
+## Training details
+Our training data is available at the HuggingFace dataset [selfrag_train_data](https://huggingface.co/datasets/selfrag/selfrag_train_data). 
+See our official repository for the training details. 
+We used 8 A100 40GB for training on the Stability HPC server.
+
 ## Citation and contact
 If you use this model, please cite our work: 
 ```
 @article{asai2023selfrag,
   author    = {Asai, Akari and Wu, Zeqiu and Wang, Yizhong and Sil, Avirup and Hajishirzi, Hannaneh},
-  title     = {{Self-RAG}:  Learning to Retrieve, Generate, and Critique through Self-Reflection},
+  title     = {{Self-RAG}: Learning to Retrieve, Generate, and Critique through Self-Reflection},
   year      = {2023},
+  journal   = { arXiv preprint arXiv:2310.11511 },
+  URL       = {https://arxiv.org/abs/2310.11511}
 }
 ```