Merge pull request #126 from 1985312383/main

Update Readme files

Author: 1985312383

.github/ISSUE_TEMPLATE/config.yml

@@ -0,0 +1,7 @@
+blank_issues_enabled: false
+
+contact_links:
+  - name: Q&A / 使用咨询
+    url: https://github.com/datawhalechina/torch-rechub/discussions
+    about: 请在 Discussions 中提出一般性问题、使用咨询或非 Bug 报告
+

README.md

@@ -14,6 +14,8 @@
 
 English | [简体中文](README_zh.md)
 
+**Online Documentation:** https://datawhalechina.github.io/torch-rechub/ (English) | https://datawhalechina.github.io/torch-rechub/zh/ (简体中文)
+
 **Torch-RecHub** is a flexible and extensible recommender system framework built with PyTorch. It aims to simplify research and application of recommendation algorithms by providing common model implementations, data processing tools, and evaluation metrics.
 
 ![Torch-RecHub Banner](docs/public/img/readme_banner_1.png)
@@ -22,7 +24,7 @@ English | [简体中文](README_zh.md)
 
 * **Modular Design:** Easy to add new models, datasets, and evaluation metrics.
 * **PyTorch-based:** Leverages PyTorch's dynamic graph and GPU acceleration capabilities.
-* **Rich Model Library:** Contains various classic and cutting-edge recommendation algorithms.
+* **Rich Model Library:** Contains various classic and cutting-edge recommendation algorithms, including matching, ranking, multi-task, and **generative models (HSTU, HLLM)**.
 * **Standardized Pipeline:** Provides unified data loading, training, and evaluation workflows.
 * **Easy Configuration:** Adjust experiment settings via config files or command-line arguments.
 * **Reproducibility:** Designed to ensure reproducible experimental results.
@@ -124,13 +126,13 @@ torch-rechub/             # Root directory
 │       └── mtl.py        # Multi-task utilities
 ├── examples/             # Example scripts
 │   ├── matching/         # Matching task examples
-│   └── ranking/          # Ranking task examples
-├── docs/                 # Documentation
+│   ├── ranking/          # Ranking task examples
+│   └── generative/       # Generative recommendation examples (HSTU, HLLM, etc.)
+├── docs/                 # Documentation (VitePress: multi-language, English & Chinese)
 ├── tutorials/            # Jupyter tutorials
 ├── tests/                # Unit tests
 ├── config/               # Configuration files
-├── scripts/              # Utility scripts
-└── mkdocs.yml            # MkDocs config file
+└── scripts/              # Utility scripts
 ```
 
 ## 💡 Supported Models
@@ -163,11 +165,17 @@ The framework currently supports the following recommendation models:
 **Multi-Task Recommendation:**
 
 * **[ESMM](https://arxiv.org/pdf/1804.07931.pdf):** Entire Space Multi-Task Model
-* **[MMoE](https://dl.acm.org/doi/pdf/10.1145/3219819.3220007):** Multi-Task Multi-Interest Network for Recommendation
+* **[MMoE](https://dl.acm.org/doi/pdf/10.1145/3219819.3220007):** Multi-gate Mixture-of-Experts for Multi-Task Learning
 * **[PLE](https://dl.acm.org/doi/pdf/10.1145/3394486.3403394):** Personalized Learning to Rank
 * **[AITM](https://arxiv.org/pdf/2005.02553.pdf):** Adaptive Interest-Task Matching
 * ... 
 
+**Generative Recommendation:**
+
+* **[HSTU](https://arxiv.org/abs/2402.17152v3):** Hierarchical Sequential Transduction Units based generative sequential recommendation model.
+* **[HLLM](https://arxiv.org/abs/2508.18118v1):** Hierarchical Large Language Model for recommendation with LLM-based item embeddings.
+* ...
+
 ## 📊 Supported Datasets
 
 The framework provides built-in support or preprocessing scripts for the following common datasets:
@@ -299,4 +307,4 @@ If you use this framework in your research or work, please consider citing:
 
 ---
 
-*Last updated: [2025-06-30]*
+*Last updated: [2025-11-27]*

README_zh.md

@@ -14,6 +14,8 @@
 
 [English](README.md) | 简体中文
 
+**在线文档：** https://datawhalechina.github.io/torch-rechub/ （英文）| https://datawhalechina.github.io/torch-rechub/zh/ （简体中文）
+
 **Torch-RecHub** 是一个使用 PyTorch 构建的、灵活且易于扩展的推荐系统框架。它旨在简化推荐算法的研究和应用，提供常见的模型实现、数据处理工具和评估指标。
 
 ![Torch-RecHub 横幅](docs/public/img/readme_banner_1.png)
@@ -22,7 +24,7 @@
 
 * **模块化设计:** 易于添加新的模型、数据集和评估指标。
 * **基于 PyTorch:** 利用 PyTorch 的动态图和 GPU 加速能力。
-* **丰富的模型库:** 包含多种经典和前沿的推荐算法（请在下方列出）。
+* **丰富的模型库:** 包含多种经典和前沿的推荐算法（召回、排序、多任务、**生成式推荐模型 HSTU/HLLM 等**）。
 * **标准化流程:** 提供统一的数据加载、训练和评估流程。
 * **易于配置:** 通过配置文件或命令行参数轻松调整实验设置。
 * **可复现性:** 旨在确保实验结果的可复现性。
@@ -126,13 +128,13 @@ torch-rechub/             # 根目录
 │       └── mtl.py        # 多任务工具
 ├── examples/             # 示例脚本
 │   ├── matching/         # 召回任务示例
-│   └── ranking/          # 排序任务示例
-├── docs/                 # 文档
+│   ├── ranking/          # 排序任务示例
+│   └── generative/       # 生成式推荐示例（HSTU、HLLM 等）
+├── docs/                 # 文档（VitePress，多语言）
 ├── tutorials/            # Jupyter教程
 ├── tests/                # 单元测试
 ├── config/               # 配置文件
-├── scripts/              # 工具脚本
-└── mkdocs.yml            # MkDocs配置文件
+└── scripts/              # 工具脚本
 ```
 
 ## 💡 支持的模型
@@ -165,11 +167,17 @@ torch-rechub/             # 根目录
 **多任务推荐 (Multi-Task Recommendation):**
 
 * **[ESMM](https://arxiv.org/pdf/1804.07931.pdf):** Entire Space Multi-Task Model
-* **[MMoE](https://dl.acm.org/doi/pdf/10.1145/3219819.3220007):** Multi-Task Multi-Interest Network for Recommendation
+* **[MMoE](https://dl.acm.org/doi/pdf/10.1145/3219819.3220007):** Multi-gate Mixture-of-Experts for Multi-Task Learning
 * **[PLE](https://dl.acm.org/doi/pdf/10.1145/3394486.3403394):** Personalized Learning to Rank
 * **[AITM](https://arxiv.org/pdf/2005.02553.pdf):** Adaptive Interest-Task Matching
 * ... 
 
+**生成式推荐 (Generative Recommendation):**
+
+* **[HSTU](https://arxiv.org/abs/2402.17152v3):** 基于层次化序列转导单元的生成式序列推荐模型。
+* **[HLLM](https://arxiv.org/abs/2508.18118v1):** 基于大语言模型表征的层次化生成式推荐模型。
+* ...
+
 ## 📊 支持的数据集
 
 框架内置了对以下常见数据集格式的支持或提供了处理脚本：
@@ -300,4 +308,4 @@ match_trainer.fit(train_dl)
 
 ---
 
-*最后更新: [2025-06-30]*
+*最后更新: [2025-11-27]*