Sungrae Park

2021.03 -. AI Research Scientist at Upstage AI (Current Position)
2018.09 - 2021.02. AI Research Scientist at CLOVA, Naver Corp.

Research Interest

• Regularization and Optimization for Deep Neural Networks

• Probabilistic Graphical Models for Data Analysis

• Analysis of Sequential Datasets

• Scene Text Detection & Recognition (OCR)

Publications

International Conference

• ZERA: Zero-prompt Evolving Refinement Agent From Zero Instructions to Structured Prompts via Principle-based Optimization (EMNLP-2025, Main)

  • Seungyoun Yi, Minsoo Khang, Sungrae Park[corresponding author]

• SRFormer: Text Detection Transformer with Incorporated Segmentation and Regression (AAAI-2024)

  • Qingwen Bu, Sungrae Park[corresponding author], Minsoo Khang, Yichuan Cheng

• Domain Generalization by Mutual-Information Regularization with Pre-trained Models. (ECCV-2022)

  • Junbum Cha, Kyungjae Lee, Sungrae Park, Sanghyuk Chun

• Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features. (ECCV-2022)

  • Byeonghu Na, Yoonsik Kim, and Sungrae Park[corresponding author]

• RewriteNet: Realistic Scene Text Image Generation via Editing Text in Real-world Image. (CVPRW-2022)

  • Junyeop Lee, Yoonsik Kim, Seonghyeon Kim, Moonbin Yim, Seung Shin, Gayoung Lee, and Sungrae Park[corresponding author]

• Contrastive Learning for Knowledge Tracing. (WWW-2022)

  • Wonsung Lee, Jaeyoon Chun, Youngmin Lee, Kyoungsoo Park, and Sungrae Park[corresponding author]

• BROS: A Pre-trained Language Model Focusing on Text and Layout for Better Key Information Extraction from Documents. (AAAI-2022)

  • Teakgyu Hong, Donghyun Kim, Mingi Ji, Wonseok Hwang, Daehyun Nam, and Sungrae Park[corresponding author]

• SWAD: Domain Generalization by Seeking Flat Minima. (NeurIPS-2021)

  • Junbum Cha, Sanghyuk Chun, Kyungjae Lee, Han-Cheol Cho, Seunghyun Park, Yunsung Lee, and Sungrae Park[corresponding author]

• SynthTIGER: Synthetic Text Image GEneratoR Towards Better Text Recognition Models. (ICDAR-2021).

  • Moonbin Yim, Yoonsik Kim, Han-Cheol Cho, and Sungrae Park[corresponding author]

• Show, Attend and Distill: Knowledge Distillation via Attention-based Feature Matching. (AAAI-2021).

  • Mingi Ji, Byeongho Heo, and Sungrae Park[corresponding author]

• Scale down Transformer by Grouping Features for a Lightweight Character-level Language Model. (COLING-2020).

  • Sungrae Park, Geewook Kim, Junyeop Lee, Junbum Cha, Ji-Hoon Kim, and Hwalsuk Lee

• Character Region Attention For Text Spotting.  (ECCV-2020).

  • Youngmin Baek, Seung Shin, Jeonghun Baek, Sungrae Park, Junyeop Lee, Daehyun Nam, and Hwalsuk Lee

• CLEval: Character-Level Evaluation for Text Detection and Recognition Tasks. (CVPRW-2020).

  • Youngmin Baek, Daehyun Nam, Sungrae Park, Junyeop Lee, Seung Shin, Jeonghun Baek, Chae-Young Lee, and Hwalsuk Lee

• On Recognizing Texts of Arbitrary Shapes with 2D Self-Attention. (CVPRW-2020).

  • Junyeop Lee, Sungrae Park, Jeonghun Baek, Seong Joon Oh, Seonghyeon Kim, and Hwalsuk Lee

• Post-OCR Parsing: Building Simple and Robust Parser via BIO Tagging. (NeurIPSW-2019). 

  • Wonseok Hwang, Seonghyeon Kim, Minjoon Seo, Jinyeong Yim, Seunghyun Park, Sungrae Park, Junyeop Lee, Bado Lee, and Hwalsuk Lee

• What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis. (ICCV-2019). 

  • Jeonghun Baek, Geewook Kim, Junyeop Lee, Sungrae Park, Dongyoon Han, Sangdoo Yun, Seong Joon Oh, and Hwalsuk Lee

• Adversarial Dropout for Recurrent Neural Networks. (AAAI-2019). 

  • Sungrae Park, Kyungwoo Song, Mingi Ji, Wonsung Lee, and Il-Chul Moon

• Hierarchical Context enabled Recurrent Neural Network for Recommendation. (AAAI-2019). 

  • Kyungwoo Song, Mingi Ji, Sungrae Park, and Il-Chul Moon

• Diagnosis Prediction via Medical Context Attention Networks Using Deep Generative Modeling. (ICDM-2018). 

  • Wonsung Lee, Sungrae Park, Weonyoung Joo, and Il-Chul Moon

• Adversarial Dropout for Supervised and SemiSupervised Learning. (AAAI-2018).

  • Sungrae Park, Jun-Keon Park, Su-Jin Shin, and Il-Chul Moon

• Hierarchical Prescription Pattern Analysis with Symptom Labels. (ICDMW-2015)

  • Su-Jin Shin, Je-Yong Oh, Sungrae Park, Minki Kim, and Il-Chul Moon

• Supervised Dynamic Topic Models for Associative Topic Extraction with A Numerical Time Series. (CIKMW-2015)

  • Sungrae Park, Wonsung Lee, and Il-Chul Moon

• Disease-Medicine Topic Models for Prescription Record Mining. (SMC-2014)

  • Sungrae Park, Doosup Choi, Wonsung Lee, Dain Jung, Minki Kim, and Il-Chul Moon

• Modeling Multiple Fields of Collective Emotions with Brownian Agents. (AAMAS-2014).

  • Wonsung Lee, Sungrae Park, and Il-Chul Moon

International Journal

• Dirichlet variational autoencoder. 

  • Weonyoung Joo, Wonsung Lee, Sungrae Park, and Il-Chul Moon

  • Pattern Recognition, 107514, 2020

• Hierarchical Prescription Pattern Analysis with Symptom Labels.

  • Su-Jin Shin, Je-Young Oh, Sungrae Park, Minki Kim, and Il-Chul Moon

  • Pattern Recognition Letters, 2018.

• Identifying Prescription Patterns with Probabilistic Graphical Model on Disease and Medicine. 

  • Sungrae Park, Doosup Choi, Minki Kim, Wonchul Cha, Chuhyun Kim, and Il-Chul Moon

  • Journal of Biomedical Informatics, 2017.

• Associative Topic Models with Numerical Time Series. 

  • Sungrae Park, Wonsung Lee, and Il-Chul Moon

  • Information Processing and Management, 2015.

• Efficient Extraction of Domain-Specific Sentiment Lexicon with Active Learning. 

  • Sungrae Park, Wonsung Lee, and Il-Chul Moon

  • Pattern Recognition Letters, 2015.

Education

Ph.D (2014.3 - 2018.7): Applied Artificial Intelligence Lab., Industrial & Systems Engineering, KAIST

M.S (2012.2 - 2014.2): Applied Artificial Intelligence Lab., Industrial & Systems Engineering, KAIST

B.S (2007.3 - 2012.2): Industrial Engineering, Hanyang University