deep-learning-analytics
commited on
Commit
•
1cc73f5
1
Parent(s):
7184940
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,39 @@
|
|
1 |
---
|
2 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
dataset:
|
3 |
+
name: arxiv_small_nougat
|
4 |
+
description: A dataset containing 108 recent papers from arXiv related to LLM (Large Language Models) and Transformers, parsed and processed using Meta's Nougat model to preserve tables and math equations.
|
5 |
+
license: [MIT]
|
6 |
+
task_categories: [Natural Language Processing, Machine Learning]
|
7 |
+
languages: [English]
|
8 |
+
size: 108 papers
|
9 |
+
download_size: [21.9MB]
|
10 |
+
|
11 |
---
|
12 |
+
|
13 |
+
## Dataset Description
|
14 |
+
|
15 |
+
The "arxiv_small_nougat" dataset is a collection of 108 recent papers sourced from arXiv, focusing on topics related to Large Language Models (LLM) and Transformers. These papers have been meticulously processed and parsed using Meta's Nougat model, which is specifically designed to retain the integrity of complex elements such as tables and mathematical equations.
|
16 |
+
|
17 |
+
## Data Format
|
18 |
+
|
19 |
+
The dataset contains the parsed content of the selected papers, with special attention given to the preservation of formatting, tables, and mathematical expressions. Each paper is provided as plain text.
|
20 |
+
|
21 |
+
## Usage
|
22 |
+
|
23 |
+
Researchers, academics, and natural language processing practitioners can leverage this dataset for various tasks related to LLM and Transformers, including:
|
24 |
+
|
25 |
+
- Language modeling
|
26 |
+
- Text summarization
|
27 |
+
- Information retrieval
|
28 |
+
- Table and equation extraction
|
29 |
+
|
30 |
+
## Acknowledgments
|
31 |
+
|
32 |
+
We acknowledge the arXiv platform for providing open access to a wealth of research papers in the field of machine learning and natural language processing.
|
33 |
+
|
34 |
+
## License
|
35 |
+
|
36 |
+
[mit]
|
37 |
+
|
38 |
+
---
|
39 |
+
|