.gitignore

+*__pycache__/
+samples*/
+runs/
+checkpoints/
+logs/
+*.DS_Store
+.idea
+output*
+test*
\ No newline at end of file

README.md

@@ -6,7 +6,7 @@
 # LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA
 
 <p align="center">
-    🤗 <a href="https://huggingface.co/datasets/THUDM/LongCite-45k" target="_blank">HF Repo</a> • 📃 <a href="https://arxiv.org/abs/2409.02897" target="_blank">Paper</a> • 🚀 <a href="https://huggingface.co/spaces/THUDM/LongCite" target="_blank">HF Space</a>
+    🤗 <a href="https://huggingface.co/datasets/THUDM/LongCite-45k" target="_blank">HF Repo</a> • 📃 <a href="https://arxiv.org/abs/2409.02897" target="_blank">Paper</a> • 🚀 <a href="https://huggingface.co/spaces/THUDM-HF-SPACE/LongCite" target="_blank">HF Space</a>
 </p>
 
 [English](./README.md) | [中文](./README_zh.md)
@@ -65,6 +65,7 @@ We are also open-sourcing CoF (Coarse to Fine) under `CoF/`, our automated SFT d
 
 You can download and save the **LongCite-45k** dataset through the Hugging Face datasets ([🤗 HF Repo](https://huggingface.co/datasets/THUDM/LongCite-45k)):
 ```python
+from datasets import load_dataset
 dataset = load_dataset('THUDM/LongCite-45k')
 for split, split_dataset in dataset.items():
     split_dataset.to_json("train/long.jsonl")

README_zh.md

@@ -5,7 +5,7 @@
 # LongCite: 让LLM在长上下文问答中生成细粒度引用
 
 <p align="center">
-    🤗 <a href="https://huggingface.co/datasets/THUDM/LongCite-45k" target="_blank">HF 库</a> • 📃 <a href="https://arxiv.org/abs/2409.02897" target="_blank">论文</a>  • 🚀 <a href="https://huggingface.co/spaces/THUDM/LongCite" target="_blank">HF空间</a>
+    🤗 <a href="https://huggingface.co/datasets/THUDM/LongCite-45k" target="_blank">HF 库</a> • 📃 <a href="https://arxiv.org/abs/2409.02897" target="_blank">论文</a>  • 🚀 <a href="https://huggingface.co/spaces/THUDM-HF-SPACE/LongCite" target="_blank">在线体验</a>
 </p>
 
 [English](./README.md) | [中文](./README_zh.md)
@@ -64,6 +64,7 @@ CUDA_VISIBLE_DEVICES=0 streamlit run demo.py --server.fileWatcherType none
 
 你可以通过 Hugging Face 数据集 ([🤗 HF Repo](https://huggingface.co/datasets/THUDM/LongCite-45k)) 下载和保存 **LongCite-45k** 数据集：
 ```python
+from datasets import load_dataset
 dataset = load_dataset('THUDM/LongCite-45k')
 for split, split_dataset in dataset.items():
     split_dataset.to_json("train/long.jsonl")