⚠️NOTE: We can't release the datasets originated from AI Hub (K-SNS, K-TDD, K-ED, K-DS) for now, according to the Terms of Use. We're in consultation with the relevant organizations and will make these public in the appropriate form soon.

Dataset Card for KoDialogBench

For most of detailed information, please refer to the following:

• Paper: KoDialogBench: Evaluating Conversational Understanding of Language Models with Korean Dialogue Benchmark
• Repository: GitHub

Dataset Details

Dataset Description

KoDialogBench is a benchmark designed to assess the conversational capabilities of language models in Korean language. To this end, we collected native Korean dialogues on daily topics from public sources (e.g., AI Hub), or translated dialogues from other languages such as English and Chinese. We then structured these conversations into diverse test datasets, spanning from dialogue comprehension to response selection tasks. This benchmark consists of 21 test sets, encompassing various aspects of open-domain colloquial dialogues (e.g., topic, emotion, dialog act).

Data Sources

We collected native Korean dialogues from AI Hub:

• K-SNS stands for Korean SNS (한국어 SNS)
• K-TDD stands for Thematic Daily Dialogues (주제별 텍스트 일상 대화 데이터)
• K-ED stands for Emotional Dialogues (감성 대화 말뭉치)
• K-DS stands for Dialogue Summary (한국어 대화 요약)

We translated public datasets from other languages:

• DailyDialog from "DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset"
• Empathetic Dialogues from "Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset"
• PersonaChat from "Personalizing Dialogue Agents: I have a dog, do you have pets too?"
• SocialDial from "SocialDial: A Benchmark for Socially-Aware Dialogue Systems"

Data Creation

We utilized diverse meta information such as dialogue topic and speaker's emotion which was already annotated in the original datasets to formulate dialogue-related multiple-choice questions. To prevent label imbalance, we sampled the equal number of examples from each class.

Statistics

The dataset has 82,962 examples in total.

Task	Subtask	Source	# Options	# Examples
Dialogue Comprehension	Topic Classification	K-SNS	6	1200
Dialogue Comprehension	Topic Classification	K-TDD	19	1900
Dialogue Comprehension	Topic Classification	SocialDial	4	400
Dialogue Comprehension	Emotion Recognition	K-ED	6	1200
Dialogue Comprehension	Emotion Recognition	DailyDialog	5	470
Dialogue Comprehension	Emotion Recognition	Empathetic Dialogues	2	2000
Dialogue Comprehension	Relation Classification	SocialDial (Distance)	4	524
Dialogue Comprehension	Relation Classification	SocialDial (Relation)	3	330
Dialogue Comprehension	Location Classification	SocialDial	4	376
Dialogue Comprehension	Dialog Act Classification	K-TDD	4	520
Dialogue Comprehension	Dialog Act Classification	DailyDialog	4	1000
Dialogue Comprehension	Fact Identification	K-DS	4	1200
Dialogue Comprehension	Fact Identification	PersonaChat	4	1000
Dialogue Comprehension	Fact Identification	Empathetic Dialogues	4	2394
Response Selection		K-SNS	5	10295
Response Selection		K-TDD	5	10616
Response Selection		K-ED	5	17818
Response Selection		PersonaChat	5	7801
Response Selection		DailyDialog	5	6740
Response Selection		Empathetic Dialogues	5	7941
Response Selection		SocialDial	5	7237

Limitations

Our benchmark may suffer from a chronic problem of benchmark contamination. Due to the scarcity of Korean language resources, there is a possibility that the held-out sources utilized to construct the benchmark might overlap with training data used for some language models.

Ethics Statement

Our benchmark dataset is designed to assess capabilities related to various situations and aspects of conversations in Korean language. To achieve this, we utilized conversational content from publicly available datasets from various sources, either without modification or with translation if necessary. During this process, there is a possibility that harmful content or inappropriate biases existing in the original data may have been conveyed, or may have arisen due to limitations of translation tools. We reject any form of violence, discrimination, or offensive language, and our benchmark dataset and experimental results does not represent such values. If any harmful content or privacy infringement is identified within the dataset, we kindly request immediate notification to the authors. In the event of such cases being reported, we will apply the highest ethical standards and take appropriate actions.

Citation

BibTeX:

@misc{jang2024kodialogbench,
      title={KoDialogBench: Evaluating Conversational Understanding of Language Models with Korean Dialogue Benchmark}, 
      author={Seongbo Jang and Seonghyeon Lee and Hwanjo Yu},
      year={2024},
      eprint={2402.17377},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}

Point of Contact

Seongbo Jang