Spaces:
Running
Running
Update src/display/about.py
Browse files- src/display/about.py +7 -11
src/display/about.py
CHANGED
@@ -17,7 +17,7 @@ class Tasks(Enum):
|
|
17 |
|
18 |
# Your leaderboard name
|
19 |
# TITLE = """<h1 align="center" id="space-title">📃 SeaExam and SeaBench Leaderboard</h1>"""
|
20 |
-
TITLE = """<h1 align="left" id="space-title">🏅 LLM Leaderboard for
|
21 |
|
22 |
# subtitle
|
23 |
SUB_TITLE = """<h2 align="left" id="space-title">What is the best LLM for Southeast Asian Languages❓</h1>"""
|
@@ -36,11 +36,12 @@ This leaderboard evaluates Large Language Models (LLMs) on Southeast Asian (SEA)
|
|
36 |
"""
|
37 |
|
38 |
INTRODUCTION_TEXT = """
|
39 |
-
This leaderboard evaluates Large Language Models (LLMs) on Southeast Asian (SEA) languages through two comprehensive benchmarks
|
40 |
-
* SeaExam assesses world knowledge and reasoning capabilities through exam-style questions [[data (public)](https://huggingface.co/datasets/SeaLLMs/SeaExam)] [[code](https://github.com/DAMO-NLP-SG/SeaExam)]
|
41 |
-
* SeaBench evaluates instruction-following abilities and multi-turn conversational skills. [[data (public)](https://huggingface.co/datasets/SeaLLMs/SeaBench)] [[code](https://github.com/DAMO-NLP-SG/SeaBench?tab=readme-ov-file)]
|
42 |
|
43 |
-
|
|
|
44 |
For more details, please refer to the "📝 About" tab.
|
45 |
"""
|
46 |
# For additional details such as datasets, evaluation criteria, and reproducibility, please refer to the "📝 About" tab.
|
@@ -119,16 +120,11 @@ If everything is done, check you can launch the EleutherAIHarness on your model
|
|
119 |
|
120 |
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
121 |
CITATION_BUTTON_TEXT = r"""
|
122 |
-
@article{damonlp2024sealeaderboard,
|
123 |
-
author = {Chaoqun Liu, Wenxuan Zhang, Jiahao Ying, Mahani Aljunied, Anh Tuan Luu, Lidong Bing},
|
124 |
-
title = {SeaExam and SeaBench: Benchmarking LLMs with Local Multilingual Questions in Southeast Asia},
|
125 |
-
year = {2024},
|
126 |
-
url = {},
|
127 |
}
|
128 |
"""
|
129 |
|
130 |
CONTACT_TEXT = f"""
|
131 |
## Contact
|
132 |
|
133 |
-
If you have any questions or want to include your models in the leaderboard, please contact Chaoqun Liu (<[email protected]>) and Wenxuan Zhang
|
134 |
"""
|
|
|
17 |
|
18 |
# Your leaderboard name
|
19 |
# TITLE = """<h1 align="center" id="space-title">📃 SeaExam and SeaBench Leaderboard</h1>"""
|
20 |
+
TITLE = """<h1 align="left" id="space-title">🏅 LLM Leaderboard for Southeast Asian Languages</h1>"""
|
21 |
|
22 |
# subtitle
|
23 |
SUB_TITLE = """<h2 align="left" id="space-title">What is the best LLM for Southeast Asian Languages❓</h1>"""
|
|
|
36 |
"""
|
37 |
|
38 |
INTRODUCTION_TEXT = """
|
39 |
+
This leaderboard evaluates Large Language Models (LLMs) on Southeast Asian (SEA) languages through two comprehensive benchmarks - SeaExam and SeaBench:
|
40 |
+
* **SeaExam** assesses world knowledge and reasoning capabilities through exam-style questions [[data (public)](https://huggingface.co/datasets/SeaLLMs/SeaExam)] [[eval code](https://github.com/DAMO-NLP-SG/SeaExam)]
|
41 |
+
* **SeaBench** evaluates instruction-following abilities and multi-turn conversational skills. [[data (public)](https://huggingface.co/datasets/SeaLLMs/SeaBench)] [[eval code](https://github.com/DAMO-NLP-SG/SeaBench?tab=readme-ov-file)]
|
42 |
|
43 |
+
Below are the aggregated results for SeaExam and SeaBench, shown both the public dataset ("pub") - which you can download via the link above - and our in-house held-out private dataset ("prv").
|
44 |
+
For detailed results by language, please refer to the individual sub-tabs.
|
45 |
For more details, please refer to the "📝 About" tab.
|
46 |
"""
|
47 |
# For additional details such as datasets, evaluation criteria, and reproducibility, please refer to the "📝 About" tab.
|
|
|
120 |
|
121 |
CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results"
|
122 |
CITATION_BUTTON_TEXT = r"""
|
|
|
|
|
|
|
|
|
|
|
123 |
}
|
124 |
"""
|
125 |
|
126 |
CONTACT_TEXT = f"""
|
127 |
## Contact
|
128 |
|
129 |
+
If you have any questions or want to include your models in the leaderboard, please contact Chaoqun Liu (<[email protected]>) and [Wenxuan Zhang](https://isakzhang.github.io/).
|
130 |
"""
|