Add paper, GitHub, and project page links; update task category
Browse filesThis PR adds links to the paper (Hugging Face papers page), the GitHub repository, and the project page in the dataset card. It also updates the `task_categories` from `text-generation` to `other` to better reflect the agent-benchmark nature of this dataset.
README.md
CHANGED
|
@@ -1,20 +1,24 @@
|
|
| 1 |
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
task_categories:
|
| 4 |
-
- text-generation
|
| 5 |
language:
|
| 6 |
- zh
|
| 7 |
- en
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
tags:
|
| 9 |
- agent
|
| 10 |
- agent benchmark
|
| 11 |
-
pretty_name: DuMateBench dataset
|
| 12 |
-
size_categories:
|
| 13 |
-
- n<1K
|
| 14 |
---
|
| 15 |
|
| 16 |
# DuMateBench Dataset
|
| 17 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
## Dataset Summary
|
| 19 |
|
| 20 |
DuMateBench is a benchmark dataset for evaluating AI agents on realistic
|
|
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
| 2 |
language:
|
| 3 |
- zh
|
| 4 |
- en
|
| 5 |
+
license: apache-2.0
|
| 6 |
+
size_categories:
|
| 7 |
+
- n<1K
|
| 8 |
+
task_categories:
|
| 9 |
+
- other
|
| 10 |
+
pretty_name: DuMateBench dataset
|
| 11 |
tags:
|
| 12 |
- agent
|
| 13 |
- agent benchmark
|
|
|
|
|
|
|
|
|
|
| 14 |
---
|
| 15 |
|
| 16 |
# DuMateBench Dataset
|
| 17 |
|
| 18 |
+
**Paper**: [DuMateBench: Evaluating Autonomous Agents in Complex Real-World Workflows](https://huggingface.co/papers/2608.26546)
|
| 19 |
+
**GitHub**: [https://github.com/baidubce/dumate-bench](https://github.com/baidubce/dumate-bench)
|
| 20 |
+
**Project Page**: [https://dumatebench.com/](https://dumatebench.com/)
|
| 21 |
+
|
| 22 |
## Dataset Summary
|
| 23 |
|
| 24 |
DuMateBench is a benchmark dataset for evaluating AI agents on realistic
|