Add model card with metadata and description

README.md

@@ -1,6 +1,18 @@
-This is the PosS-1 model of the paper **PosS:Position Specialist Generates Better Draft for Speculative Decoding**
+---
+pipeline_tag: text-generation
+library_name: transformers
+license: apache-2.0
+---
 
-If the code fails to auto-download the models, you may mannually download the following files.
+# PosS: Position Specialist Generates Better Draft for Speculative Decoding
 
-- `pytorch_model.bin`: Model weights
-- `config.json`: Model config
+This model, presented in [POSS: Position Specialist Generates Better Draft for Speculative Decoding](https://huggingface.co/papers/2506.03566), improves speculative decoding in Large Language Models (LLMs). PosS utilizes multiple position-specialized draft layers to generate tokens, mitigating error accumulation and improving the acceptance rate of later-position tokens.
+
+**Key Features:**
+
+* Position Specialists for improved token prediction accuracy at all positions.
+* Enhanced average acceptance length and speed-up ratio compared to baseline methods.
+
+**Code:** [https://github.com/shrango/PosS](https://github.com/shrango/PosS)
+
+For detailed usage instructions, evaluation methods, and training details, please refer to the GitHub repository.  Pre-trained weights are available for Llama-3-8B-Instruct and Llama-2-13B-chat.