Add pipeline tag and paper link

README.md

@@ -1,24 +1,27 @@
 ---
+base_model: Qwen/Qwen3-0.6B
+language:
+- zh
 library_name: transformers
 license: apache-2.0
-language:
-  - zh
-base_model: Qwen/Qwen3-0.6B
+pipeline_tag: text-classification
 tags:
-  - qwen3
-  - custom_code
-  - turn-taking
-  - turn-detection
-  - spoken-dialogue
-  - duplex-conversation
-  - conversational
-  - vllm
+- qwen3
+- custom_code
+- turn-taking
+- turn-detection
+- spoken-dialogue
+- duplex-conversation
+- conversational
+- vllm
 ---
 
 **English** | [中文](./README_zh.md)
 
 # XTurnix Pretrained (Qwen3 0.6B)
 
+This model is described in [XTurnix: Large-Scale Self-Supervised Turn Control through Two-State Binary Decisions](https://huggingface.co/papers/2610.04400).
+
 [![GitHub](https://img.shields.io/badge/GitHub-Repository-181717?logo=github)](https://github.com/xcc-zach/xturnix)
 
 Predict turn-taking decisions from transcribed dialogue history: whether the AI
@@ -112,4 +115,4 @@ Use the included `vllm_client.py` as the reference client.
 ## Complete Deployment
 
 This model is supported by the
-[X-Talk dialogue framework](https://github.com/xcc-zach/xtalk).
+[X-Talk dialogue framework](https://github.com/xcc-zach/xtalk).