한밭대학교컴퓨터공학과

Knowledge Intelligence Lab

About the Lab

The Knowledge Intelligence Lab focuses on natural language processing, which is a branch of artificial intelligence. Natural language processing is the study of enabling human-machine communication. Our lab applies learning methods to analyze and generate meaningful information from a wide variety of data in the world.

Main Research Areas

□ Natural language processing research: morphological analysis, part-of-speech tagging, object name recognition, syntactic analysis, semantic analysis, and discourse analysis

□ Natural language processing applications: Q&A, machine translation, document summarization, text generation, conversational AI (Agent)

□ Modern natural language processing: RAG and LLM, NLP-based multimodal, reasoning, NLP security, NLP for robotics

□ MLOps: TorchServe, Prefect, BentoML

* Abbreviations

- NLP: Natural Language Processing

- RAG: Retrieval-Augmented Generation

- LLM: Large Language Model

Advisor: Cheon Eum Park

Thesis

• ADMit: Improving NER in Automotive Domain with Domain Adversarial Training and Multi-task Learning, ESWA, Sep. 2023
• Robust Multi-task Learning-based Korean POS Tagging to Overcome Word Spacing Errors, TALLIP, Jun. 2023
• Multi-task Learning with Contextual Hierarchical Attention for Korean Coreference Resolution, ETRIJ, Feb. 2022
• CrossAug: A Contrastive Data Augmentation Method for Debiasing Fact Verification Models, CIKM, Oct. 2021
• Simple and Effective Neural Coreference Resolution for Korean, ETRIJ, Dec. 2021
• Korean TableQA: Structured Data Question Answering based on Span Prediction Style with S^3–NET, ETRIJ, Dec. 2020
• Fast End-to-end Coreference Resolution for Korean, EMNLP, Nov. 2020
• VS^3-NET: Neural Variational Inference for Machine Reading Comprehension, ETRIJ, Dec. 2019
• CoNLL 2018 Shared Task, SEx BiST: A Multi-Source Trainable Parser with Deep Contextualized Lexical Representations (2nd UAS, 4th LAS, first place of parsing English, French and Korean), Nov. 2018

Patents

• METHOD OF TRAINING POS TAGGING MODEL, COMPUTER-READABLE RECORDING MEDIUM AND POS TAGGING METHOD. Feb 01, 2024, US: 18/205,615
• SYSTEM FOR GENERATING ANSWERS BASED ON MULTITASK LEARNING AND CONTROL METHOD THEREOF. Nov 16, 2023, US: 18/195,770
• METHOD FOR TRAINING SLOT TAGGING MODEL, COMPUTER-READABLE MEDIUM, SPEECH RECOGNITION APPARATUS AND ELECTRONIC DEVICE. Sep 14, 2023, US: 18/081,464
• NAMED ENTITY RECOGNITION SYSTEM AND NAMED ENTITY RECOGNITION METHOD. Jun 29, 2023, US: 18/065,891
• QUESTION AND ANSWER SYSTEM AND METHOD FOR CONTROLLING THE SAME. 2020/10, CN: 202210149110.9, US: 17/673,287
• KOREAN MORPHOLOGICAL ANALYSIS METHOD WITH LINEAR TIME COMPLEXITY USING A RESTRICTED NEURAL NETWORK AND RECORDING MEDIUM RECORDING THE SAME). 2019/11/21, 10-2049517 (REGISTERED)
• CROSS-REFERENCE RESOLUTION SYSTEM AND METHOD USING A HIERARCHICAL POINTER NETWORK. 2019/10/11, 10-2033458 (REGISTERED)
• KOREAN DEPENDENT PHRASE ANALYSIS SYSTEM AND METHOD USING A MULTI-TASK LEARNING-BASED POINTER NETWORK. 2019/02/11, 10-1948613 (REGISTERED)