Datasets:

Modalities:
Text
Formats:
csv
Languages:
Thai
Libraries:
Datasets
Dask
License:
File size: 3,843 Bytes
0718122
 
 
 
 
 
 
 
 
 
 
 
 
 
2197ece
c69c9f1
0718122
 
2197ece
 
 
 
 
c69c9f1
2197ece
 
 
56f10b2
 
 
 
 
 
0718122
 
 
 
0e03c05
0718122
28f600a
 
7298988
 
28f600a
 
 
7298988
 
 
28f600a
 
8fe131d
 
 
 
 
 
 
 
 
 
 
 
3b0baf1
 
 
 
8fe131d
 
 
 
 
3b0baf1
8fe131d
 
20f1e18
 
9d12210
20f1e18
 
 
2b9e310
1bda3f3
05124b3
e166879
5b46212
28f600a
0718122
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
---
license: apache-2.0
task_categories:
- question-answering
language:
- th
tags:
- bhuddist
- exam
- thai
---

# Thai Buddhist Studies Examination (Nak Tham)

This repository contains multiple-choice questions from the Thai Buddhist Studies
(Nak Tham) examination (2020, 2022, 2023). This dataset can be used for a benchmark for evaluating Large Language Models'
understanding of Thai Buddhist concepts and teachings.


## Dataset Statistics

| Year | Number of Multiple Choice Questions |
|------|-------------------------------------|
| 2020 | 1,350                              |
| 2022 | 1,400                              |
| 2023 | 1,350                              |

**Phra Udom thought on the exam:** We have reviewed the Nak Tham examination format and find it to be well-designed.
While it maintains a multiple-choice format, the questions go beyond mere memorization.
Instead, they require logical reasoning and inference skills to arrive at the correct answers.
The questions test the understanding of Buddhist principles through analytical thinking rather than rote learning.


## Data Collection Process

1. **OCR Extraction**: Questions were initially extracted from PDF examination papers using Optical Character Recognition (OCR)
2. **Data Structuring**: All questions were compiled and structured into a standardized CSV format
3. **Manual Verification**: The dataset underwent manual review and correction to ensure accuracy of text, choices, and correct answers

## Usage

You can use `datasets` to download exam in all years as shown below or download individual CSV file from `data` folder.

```py
from datasets import load_dataset

exam_dataset = load_dataset("biodatlab/thai_buddhist_studies_exam")  # all exam
exam_dataset_2023 = exam_dataset.filter(lambda e: e["year"] == 2023)  # filter only the latest year
exam_dataset_2023["train"][0]  # no data splitting, only train
```

```py
# Example prompt template to answer the question
q = exam_dataset["train"][0]
prompt = f"""As a Buddhist scholar, please help answer questions about Buddhist principles and practices.
Given the following question and choices, provide:
1. The correct answer
2. A detailed explanation of why this is correct
3. References to Buddhist texts or principles if applicable

Question: {q["question"]}

Choices:
ก. {q['a']}
ข. {q['b']}
ค. {q['c']}
ง. {q['d']}

Please structure your response in Thai language in JSON format as follows
  คำตอบ: [correct choice]
  คำอธิบาย: [explanation]
"""
# check and evaluate the answer ...
```

## Preliminary Evaluation

We calculate model accuracy of 2023 exam based on exam levels and get the following accuracy (%):

| Model | All | Elementary | Intermediate | Advanced |
|-------|-----|------------|--------------|-----------|
| Gemini-1.5-Flash | 67.93 | 66.22 | 65.78 | 71.78 |
| GPT-4o | 86.74 | 86.22 | 85.56 | 88.44 |
| Claude-3.5-Sonnet | 84.96 | 80.89 | 84.22 | 89.78 |
| Deepseek V3 | 85.18 | 84.89 | 82.89 | 87.78 |
| Deepseek R1 | 89.11 | 86.56 | 88.89 | 90.89 |

## About Dhamma Practice Exam

The Nak Tham examination is a system for evaluating Buddhist knowledge levels among ordained monks
and Buddhist practitioners. It is organized by the Religious Affairs Department's Dhamma Education Office
(Sanam Luang). The examination is held once annually and is divided into two main categories:

1. "Nak Tham" - Examinations for Buddhist monks and novices
2. "Dhamma Studies" (Tham Sueksa) - Examinations for laypeople and general public interested in Buddhist teachings

Both categories are structured into three levels:
- Elementary Level (Chan Tri)
- Intermediate Level (Chan Tho)
- Advanced Level (Chan Ek)

## Contributors

- (Phra) Udom Siritientong, Fudan University
- Titipat Achakulvisut, Department of Biomedical Engineering, Mahidol University