Update README.md
Browse files
README.md
CHANGED
|
@@ -7,6 +7,39 @@ base_model:
|
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
---
|
| 9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
The official model weight for RAG Error-Critic Model (3B), which used to identify the error response of RAG models. Here we illustrate the input-output format for our model:
|
| 11 |
|
| 12 |
|
|
|
|
| 7 |
pipeline_tag: text-generation
|
| 8 |
---
|
| 9 |
|
| 10 |
+
### <div align="center">RAG-Critic: Leveraging Automated Critic-Guided Agentic Workflow for Retrieval Augmented Generation<div>
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
|
| 15 |
+
<p align="center">
|
| 16 |
+
<br>
|
| 17 |
+
<a href="https://scholar.google.com/citations?user=amozZDkAAAAJ&hl=zh-CN" target="_blank">Guanting Dong</a>,
|
| 18 |
+
<a href="https://scholar.google.com/citations?user=iHCuEo8AAAAJ&hl=zh-CN" target="_blank">Jiajie Jin</a>,
|
| 19 |
+
<a href="https://scholar.google.com/citations?user=XDljV4YAAAAJ&hl=zh-CN" target="_blank">Xiaoxi Li</a>,
|
| 20 |
+
<a href="https://scholar.google.com/citations?user=tBqVOWsAAAAJ&hl=zh-CN" target="_blank">Yutao Zhu</a>,
|
| 21 |
+
<a href="http://playbigdata.ruc.edu.cn/dou/" target="_blank">Zhicheng Dou</a> <sup>✉</sup>;
|
| 22 |
+
<a href="https://scholar.google.com/citations?user=tbxCHJgAAAAJ&hl=en" target="_blank">Ji-rong Wen</a>
|
| 23 |
+
|
| 24 |
+
<br>
|
| 25 |
+
</p>
|
| 26 |
+
|
| 27 |
+
<p align="center">
|
| 28 |
+
<br>
|
| 29 |
+
<a href="https://github.com/dongguanting/RAG-Critic" target="_blank">Github Page</a>,
|
| 30 |
+
<br>
|
| 31 |
+
</p>
|
| 32 |
+
|
| 33 |
+
<p align="center">
|
| 34 |
+
|
| 35 |
+
Gaoling School of Artificial Intelligence, Renmin University of China.
|
| 36 |
+
<br>
|
| 37 |
+
<sup>✉</sup> Corresponding Author
|
| 38 |
+
</p>
|
| 39 |
+
|
| 40 |
+
|
| 41 |
+
|
| 42 |
+
|
| 43 |
The official model weight for RAG Error-Critic Model (3B), which used to identify the error response of RAG models. Here we illustrate the input-output format for our model:
|
| 44 |
|
| 45 |
|