I am a postdoc researcher in BK21 FOUR R&E Center for Artificial Intelligence at Korea University. I have obtained my Ph.D. in Computer Science and Engineering from Korea University, advised by Prof. SangKeun Lee. I received B.S. in Computer Science and Engineering from Korea University.
My primary research lies in the area of natural language processing and efficient AI:
- Commonsense Reasoning: Improving the commonsense reasoning capabilities of pre-trained models including language models and multi-modal transformers. Interested in augmenting the models with (implicit) commonsense knowledge by incorporating knowledge graphs and retrieving texts.
- Language Models: Improving the various aspects of language models including effectiveness, efficiency, diversity, multilinguality, etc. Addressing any pipeline of language model development, ranging from pre-training to inference.
- Model Compression: Compressing and accelerating pre-trained models for efficient inference, such as network pruning, quantization, and knowledge distillation.
Education
-
Korea University
Ph.D. in Computer Science and Engineering
Adviser: Prof. SangKeun Lee
Mar 2018 - Feb 2023 -
Korea University
B.S. in Computer Science and Engineering
Mar 2014 - Feb 2018
Experience
-
BK21 FOUR R&E Center for Artificial Intelligence, Korea University
Postdoc researcher
Mar 2023 - Present
Publications
2024
-
Coconut: Contextualized Commonsense Unified Transformers for Graph-Based Commonsense Augmentation of Language Models
Jun-Hyung Park, Mingyu Lee, Junho Kim, and SangKeun Lee
ACL 2024 Under Review -
Let’s Combine KOPS: Phoneme-Guided Korean Out-of-Vocabulary Processing System
Nayeon Kim, Eojin Jeon, Jun-Hyung Park, Mingyu Kim, and SangKeun Lee
ACL 2024 Under Review
2023
-
DIVE: Towards Descriptive and Diverse Visual Commonsense Generation
Jun-Hyung Park*, Hyuntae Park*, Youjin Kang, Eojin Jeon, and SangKeun Lee
Proceedings of EMNLP 2023 -
Leap-of-Thought: Accelerating Transformers via Dynamic Token Routing
Yeachan Kim, Junho Kim, Jun-Hyung Park, Mingyu Lee, and SangKeun Lee
Proceedings of EMNLP 2023 -
SMoP: Towards Efficient and Effective Prompt Tuning with Sparse Mixture-of-Prompts
Joon-Young Choi, Junho Kim, Jun-Hyung Park, Wing-Lam Mok, and SangKeun Lee
Proceedings of EMNLP 2023 -
Client-Customized Adaptation for Parameter-Efficient Federated Learning
Yeachan Kim*, Junho Kim*, Wing-Lam Mok, Jun-Hyung Park and SangKeun Lee
Findings of ACL 2023 -
Dynamic Structure Pruning for Compressing CNNs
Jun-Hyung Park, Yeachan Kim, Junho Kim, Joon-Young Choi, and SangKeun Lee
Proceedings of AAAI 2023
2022
-
Break it Down into BTS: Basic, Tiniest Subword Units for Korean
Jun-Hyung Park*, Nayeon Kim*, Joon-Young Choi, Eojin Jeon, Youjin Kang, and SangKeun Lee
Proceedings of EMNLP 2022 -
Tutoring Helps Students Learn Better: Improving Knowledge Distillation for BERT with Tutor Network
Jun-Hyung Park*, Junho Kim*, Mingyu Lee, Wing-Lam Mok, Joon-Young Choi, and SangKeun Lee
Proceedings of EMNLP 2022 -
Efficient Pre-training of Masked Language Model via Concept-based Curriculum Masking
Jun-Hyung Park*, Mingyu Lee*, Junho Kim, Kang-Min Kim, and SangKeun Lee
Proceedings of EMNLP 2022 -
Quantized Sparse Training: A Unified Trainable Framework for Joint Pruning and Quantization of DNNs
Jun-Hyung Park, Kang-Min Kim, and SangKeun Lee
ACM TECS Volume 21, Issue 5 -
Learning from Missing Relations: Contrastive Learning with Commonsense Knowledge Graphs for Commonsense Inference
Jun-Hyung Park*, Yong-Ho Jung*, Joon-Young Choi, Mingyu Lee, Junho Kim, Kang-Min Kim, and SangKeun Lee
Findings of ACL 2022 -
Examining the Impact of Adaptive Convolution on Natural Language Understanding
Jun-Hyung Park, Byung-Ju Choi, and SangKeun Lee
ESWA Volume 189
2021
-
KOAS: Korean Text Offensiveness Analysis System
San-Hee Park*, Kang-Min Kim*, Seonhee Cho*, Jun-Hyung Park, Hyuntae Park, Hyuna Kim, Seongwon Chung, and SangKeun Lee
Proceedings of EMNLP 2021 (Demo)
2020
-
Multi-pretraining for Large-scale Text Classification
Kang-Min Kim, Bumsu Hyeon, Yeachan Kim, Jun-Hyung Park, and SangKeun Lee
Findings of EMNLP 2020
2019
-
meChat: In-device Conversational Photo Sharing Service (demo)
Jungho Lee, Woo-Jong Ryu, Yoonjoo Ahn, Song-Eun Lee, Kang-Min Kim, Jun-Hyung Park, SangKeun Lee
Proceedings of MobiSys 2019 -
Adaptive Convolution for Text Classification
Byung-Ju Choi, Jun-Hyung Park, and SangKeun Lee
Proceedings of NAACL 2019
2018
-
meChat: In-Device Personal Assistant for Conversational Photo Sharing
Kang-Min Kim*, Woo-Jong Ryu*, Jun-Hyung Park, SangKeun Lee
IEEE IC Volume 23, Issue 2
Teaching
-
DFE624 Introduction to Data Science in Python
Instructor
Fall 2023 at Korea University -
DFE611 Modern Information Retrieval Techniques
Instructor
Spring 2023 at Korea University
Service
-
Area Chair
- ACL (2024)
-
Program Committee
- ACL (2023)
- AAAI (2023, 2024)
- EMNLP (2022, 2023)
- NAACL (2024)
- IEEE Transactions on Neural Networks and Learning Systems (2024)
- International Journal of Machine Learning and Cybernetics (2024)