Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
<h1>WebGLM: Towards An Efficient Web-enhanced Question Answering System with Human Preference</h1>
|
| 2 |
+
|
| 3 |
+
<p align="center">
|
| 4 |
+
📃 <a href="https://arxiv.org/pdf/2306.07906.pdf" target="_blank">Paper (KDD 2023)</a>
|
| 5 |
+
|
|
| 6 |
+
💻 <a href="https://github.com/THUDM/WebGLM" target="_blank">Github Repo</a>
|
| 7 |
+
</p>
|
| 8 |
+
|
| 9 |
+
# Introduction
|
| 10 |
+
|
| 11 |
+
WebGLM aspires to provide an efficient and cost-effective web-enhanced question-answering system using the 10-billion-parameter General Language Model (GLM). It aims to improve real-world application deployment by integrating web search and retrieval capabilities into the pre-trained language model.
|
| 12 |
+
|
| 13 |
+
WebGLM is built by the flowing parts:
|
| 14 |
+
|
| 15 |
+
- **LLM-augmented Retriever**: Enhances the retrieval of relevant web content to better aid in answering questions accurately.
|
| 16 |
+
- **Bootstrapped Generator**: Generates human-like responses to questions, leveraging the power of the GLM to provide refined answers.
|
| 17 |
+
- **Human Preference-aware Scorer**: Estimates the quality of generated responses by prioritizing human preferences, ensuring the system produces useful and engaging content.
|
| 18 |
+
|
| 19 |
+
This repo is the implementation of **Bootstrap Generator**.
|
| 20 |
+
|
| 21 |
+
See our [Github Repo](https://github.com/THUDM/WebGLM) for more detailed usage.
|