Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
json
Languages:
Chinese
Size:
10K - 100K
ArXiv:
License:
Add link to paper, Github repository, ethics statement and acknowledgement.
#2
by
nielsr
HF Staff
- opened
README.md
CHANGED
@@ -1,25 +1,24 @@
|
|
1 |
---
|
2 |
-
license: cc-by-nc-4.0
|
3 |
-
task_categories:
|
4 |
-
- text-classification
|
5 |
language:
|
6 |
- zh
|
|
|
7 |
size_categories:
|
8 |
- 10K<n<100K
|
|
|
|
|
9 |
---
|
|
|
10 |
<h1 align="center"> ChineseHarm-bench</h1>
|
11 |
<h3 align="center"> A Chinese Harmful Content Detection Benchmark </h3>
|
12 |
|
13 |
> ⚠️ **WARNING**: This project and associated data contain content that may be toxic, offensive, or disturbing. Use responsibly and with discretion.
|
14 |
|
15 |
<p align="center">
|
16 |
-
<a href="">Project</a> •
|
17 |
-
<a href="">Paper</a> •
|
18 |
<a href="https://huggingface.co/collections/zjunlp/chineseharm-bench-683b452c5dcd1d6831c3316c">Hugging Face</a>
|
19 |
</p>
|
20 |
|
21 |
-
|
22 |
-
|
23 |
<div>
|
24 |
</div>
|
25 |
<div align="center">
|
@@ -39,6 +38,19 @@ Each file is a list of examples with:
|
|
39 |
* `"文本"`: the input Chinese text
|
40 |
* `"标签"`: the ground-truth label
|
41 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
## 🚩Citation
|
43 |
|
44 |
Please cite our repository if you use ChineseHarm-bench in your work. Thanks!
|
|
|
1 |
---
|
|
|
|
|
|
|
2 |
language:
|
3 |
- zh
|
4 |
+
license: cc-by-nc-4.0
|
5 |
size_categories:
|
6 |
- 10K<n<100K
|
7 |
+
task_categories:
|
8 |
+
- text-classification
|
9 |
---
|
10 |
+
|
11 |
<h1 align="center"> ChineseHarm-bench</h1>
|
12 |
<h3 align="center"> A Chinese Harmful Content Detection Benchmark </h3>
|
13 |
|
14 |
> ⚠️ **WARNING**: This project and associated data contain content that may be toxic, offensive, or disturbing. Use responsibly and with discretion.
|
15 |
|
16 |
<p align="center">
|
17 |
+
<a href="https://github.com/zjunlp/ChineseHarm-bench">Project</a> •
|
18 |
+
<a href="https://arxiv.org/abs/2506.10960">Paper</a> •
|
19 |
<a href="https://huggingface.co/collections/zjunlp/chineseharm-bench-683b452c5dcd1d6831c3316c">Hugging Face</a>
|
20 |
</p>
|
21 |
|
|
|
|
|
22 |
<div>
|
23 |
</div>
|
24 |
<div align="center">
|
|
|
38 |
* `"文本"`: the input Chinese text
|
39 |
* `"标签"`: the ground-truth label
|
40 |
|
41 |
+
## 🚩 Ethics Statement
|
42 |
+
|
43 |
+
We obtain all data with proper authorization from the respective data-owning organizations and signed the necessary agreements.
|
44 |
+
|
45 |
+
**The benchmark is released under the CC BY-NC 4.0 license.
|
46 |
+
All datasets have been anonymized and reviewed by the Institutional Review Board (IRB) of the data provider to ensure privacy protection.**
|
47 |
+
|
48 |
+
Moreover, we categorically denounce any malicious misuse of this benchmark and are committed to ensuring that its development and use consistently align with human ethical principles.
|
49 |
+
|
50 |
+
## Acknowledgement
|
51 |
+
|
52 |
+
We gratefully acknowledge Tencent for providing the dataset and LLaMA-Factory for the training codebase.
|
53 |
+
|
54 |
## 🚩Citation
|
55 |
|
56 |
Please cite our repository if you use ChineseHarm-bench in your work. Thanks!
|