File size: 1,586 Bytes
52e6e50
 
 
 
 
 
 
 
 
 
 
f178a27
1fddf39
f85139f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
52e6e50
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
---
license: apache-2.0
task_categories:
- image-to-text
language:
- en
tags:
- OCR
- KIE
- Latex
- code
size_categories:
- 10K<n<100K
---

# Latex-KIE Dataset

The **Latex-KIE** dataset is a large-scale collection of paired LaTeX formula images and their corresponding LaTeX code. It is specifically designed for training and evaluating models for **Image-to-LaTeX**, **Key Information Extraction (KIE)**, and **Optical Character Recognition (OCR)** tasks in scientific domains.

---

## 📊 Dataset Summary

- **Images**: Rendered LaTeX math formulas (black text on white background)
- **Text**: Corresponding raw LaTeX code for each image
- **Split**: `train`
- **Total Samples**: 92,057
- **Format**: Parquet (`.parquet`)
- **Size**: ~439 MB

---

## 🧾 Data Fields

Each data sample consists of:

| Column         | Type       | Description                              |
|----------------|------------|------------------------------------------|
| `image`        | Image      | Rendered image of the LaTeX formula      |
| `latex_formula`| `string`   | Corresponding LaTeX string representation|

---

## 📂 Example

```json
{
  "image": "<Rendered Image of LaTeX>",
  "latex_formula": "\\begin{align*} L_{N,M,N} = \\frac{1}{N^d} \\sum ... \\end{align*}"
}
```

---

## 🧠 Use Cases

This dataset is intended for:

- Training models for **Image-to-LaTeX generation**
- Key Information Extraction (KIE) from scientific formulas
- Benchmarking OCR models on scientific/math notation
- Pretraining/fine-tuning Transformer or CNN-based encoders for math-to-text generation

---