Update model card: correct pipeline tag, add library name and paper/code links

README.md

@@ -1,14 +1,18 @@
 ---
-license: apache-2.0
+base_model: hfl/chinese-electra-180g-base-discriminator
 language:
 - zh
-base_model:
-- google/electra-base-discriminator
-pipeline_tag: text-classification
+license: apache-2.0
+pipeline_tag: token-classification
+library_name: transformers
 ---
 
 # 中文检错前筛模型：背景、训练、改造与结果摘要
 
+This model is a character-level Chinese error detector presented in the paper [CSRP: Chain-of-Thought Reasoning for Chinese Text Correction via Reinforcement Learning with Efficiency-Aware Rewards](https://huggingface.co/papers/2606.00020). 
+
+The official code is available in the [ChineseErrorCorrector](https://github.com/TW-NLP/ChineseErrorCorrector) repository.
+
 本文档概括本仓库在「纠错前筛」方向上的**字级**检测目标、数据与上游来源、训练与改造过程，以及阶段性评测与性能结论。操作级命令仍以 `UPLOAD_AND_TRAIN.txt`、`CHAR_PIPELINE.txt` 为准。
 
 ---
@@ -177,3 +181,15 @@ python download_base_model.py --model_id hfl/chinese-electra-180g-base-discrimin
 ## 8. 引用与致谢
 
 方法与数据谱系主要参考 **TW-NLP** 的 **ChineseErrorCorrector** 项目及其公开的 Hugging Face 数据集与模型卡片；预训练骨干为 **哈工大 HFL** 发布的 **Chinese ELECTRA**（discriminator）。具体论文与引用格式见上游与 [ELECTRA 模型页](https://huggingface.co/hfl/chinese-electra-180g-base-discriminator) 相关说明。
+
+```bibtex
+@misc{tian2026csrpchainofthoughtreasoningchinese,
+      title={CSRP: Chain-of-Thought Reasoning for Chinese Text Correction via Reinforcement Learning with Efficiency-Aware Rewards}, 
+      author={Wei Tian and Yuhao Zhou and Man Lan},
+      year={2026},
+      eprint={2606.00020},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL},
+      url={https://arxiv.org/abs/2606.00020}, 
+}
+```