Publications

Recent research outputs

International Journal Domestic Journal International Conference Domestic Conference

2025

MIDAS: Multi-level Intent, Domain, And Slot Knowledge Distillation for Multi-turn NLU

Yan Li, So-Eon Kim, Seong-Bae Park, Soyeon Caren Han

Findings of the Association for Computational Linguistics: NAACL 2025, pp. 8004–8027, 2025.

2024

Transformers with Attentive Federated Aggregation for Time Series Stock Forecasting

Chu Myaet Thwal, Ye Lin Tun, Kitae Kim, Seong-Bae Park and Choong Seon Hong

Proceedings of IEEE International Conference on Information Networking (ICOIN), pp. 499-504, 2024.

2023

Edge-assisted Attention-based Federated Learning for Multi-Step EVSE-enabled Prosumer Energy Demand Prediction

Luyao Zou, Chu Myaet Thwal, Seong-Bae Park, and Choong Seon Hong

Proceedings of the International Conference on Information Networking (ICOIN), pp. 116-121, 2023.

Federated Learning with Intermediate Representation Regularization

Ye Lin Tun, Chu Myaet Thwal, Yu Min Park, Seong-Bae Park and Choong Seon Hong

Proceedings of IEEE International Conference on Big Data and Smart Computing (BigComp), pp. 56-63, 2023.

Query Similarity of Various Linguistic Levels for Hybridized Conversational Agents

So-Eon Kim, Choong Seon Hong, and Seong-Bae Park

Proceedings of the 2023 International Conference on Electronics, Information, and Communication (ICEIC), pp. 1–4, 2023.

2022

An Explainable Artificial Intelligence Framework for Quality-Aware IoE Service Delivery

Md. Shirajum Munir, Seong-Bae Park, and Choong Seon Hong

Proceedings of the IEEE International Conference on Communications (ICC), pp. 4787-4793, 2022.

Question Difficulty Estimation with Directional Modality Association in Video Question Answering

Bong-Min Kim and Seong-Bae Park

In Advances and Trends in Artificial Intelligence. Theory and Practices in Artificial Intelligence (EAAI 2022), Lecture Notes in Computer Science, vol. 13343, 2022.

Post-training with Data Augmentation for Improving T5-based Question Generator

Gyu-Min Park, Seong-Eun Hong, Choong Seon Hong, and Seong-Bae Park

In Advances in Computer Science and Ubiquitous Computing (CUTECSA 2022), Lecture Notes in Electrical Engineering, vol. 1028, 2023.

A Memory Replay based Continual Learning Utilizing Class Representative and Class Boundary Data

Jun-Beom Lee, Choong Seon Hong, and Seong-Bae Park

In Advances in Computer Science and Ubiquitous Computing (CUTECSA 2022), Lecture Notes in Electrical Engineering, vol. 1028, 2023.

Post-Training with Interrogative Sentences for Enhancing BART-based Korean Question Generator

Gyu-Min Park, Seong-Eun Hong, and Seong-Bae Park

Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (IJCNLP-AACL), pp. 202-209, 2022.

2021

CbI-M: Context-based Intervention Methodology for Rehabilitation of Persons with NCDs

Hafiz Syed Muhammad Bilal, Usman Akhtar, Asim Abbas, Muhammad Asif Razzaq, Ubaid Ur Rehman, Jamil Hussain, Sun Moo Kang, Sungyoung Lee, and Seong-Bae Park

Proceedings of the 15th International Conference on Ubiquitous Information Management and Communication (IMCOM), pp. 1-5, 2021.

2020

WIRE: An Automated Report Generation System using Topical and Temporal Summarization

Yunseok Noh, Yongmin Shin, Junmo Park, A.-Yeong Kim, Su-Jeong Choi, Hyun-Je Song, Seong-Bae Park, and Se-Young Park

Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 2169–2172, 2020.

2019

Robust Sentence Classification by Solving Out-of-Vocabulary Problem with Auxiliary Word Predictor

Sang-Seok Park, Yunseok Noh, Seyoung Park, and Seong-Bae Park

Proceedings of the 16th Pacific Rim International Conferences on Artificial Intelligence (PRICAI), pp. 338-350, 2019.

2018

An End-to-End Dialog State Tracking for Task-Oriented Dialogs

A-Yeong Kim, Tae-Hyeong Kim, Hyun-Je Song, and Seong-Bae Park

Proceedings of the 2018 IEEE International Conference on Fuzzy Systems, pp. 84-90, 2018.

Publications

Low-Resourced Alphabet-Level Pivot-Based Neural Machine Translation for Translating Korean Dialects

Position-based focal loss for diverse and relevant response generation

A Digital Processing in Memory Architecture Using TCAM for Rapid Learning and Inference Based on a Spike Location Dependent Plasticity

Neuro-Symbolic Explainable Artificial Intelligence Twin for Zero-Touch IoE in Wireless Network

Token-Level Fact Correction in Abstractive Summarization

A Response Generator with Response-Aware Encoder for Generating Specific and Relevant Responses

RoBERTa-Based Keyword Extraction from Small Number of Korean Documents

Question Difficulty Estimation via Enhanced Directional Modality Association Transformer

A Non-Autoregressive Neural Machine Translation Model With Iterative Length Update of Target Sentence

Strong Influence of Responses in Training Dialogue Response Generator

Question Difficulty Estimation Based on Attention Model for Question Answering

Secure Health Fog: A Novel Framework for Personalized Recommendations Based on Adaptive Model Tuning

Enriching Knowledge Base by Parse Tree Pattern and Semantic Filter

A Topical Category-Aware Neural Text Summarizer

Learning Translation-Based Knowledge Graph Embeddings by N-Pair Translation Loss

An Approach to Knowledge Base Completion by a Commitee-Based Knowledge Graph Embedding

Korean Part-of-Speech Tagging Based on Morpheme Generation

Aspect-Based Sentiment Analysis Using Aspect Map

Abstractive Sentence Compression with Event Attention

Identifying Intention Posts in Discussion Forums Using Multi-Instance Learning and Multiple Sources Transfer Learning

A Mutlitask-Based Neural Machine Translation Model with Part-of-Speech Tags Integration for Arabic Dialects

Resolving Ambiguity in Visual Question Answering through an Iterative Clarifying QA-based Framework

Response-aware Retrieving Candidate Pair for Dialogue Generation

Post-training Methods for Improving Korean Document Summarization Model

Multi-Document Summarization Use Semantic Similarity and Information Quantity of Sentence

Response-Considered Query Token Importance Weight Calculator with Potential Response for Generating Query-Relevant Responses

Confident Multiple Choice Learning-based Ensemble Model for Video Question-Answering

Deletion-Based Sentence Compression Using Sentence Scoring Reflecting Linguistic Information

Multi-Domain Response Generation Using Memory-Replay Continual Learning

Topic Centric Korean Text Summarization using Attribute Model

Scene Generation from a Sentence by Learning Object Relation

Solving for Redundant Repetition Problem of Generating Summarization using Decoding History

Regularizing Korean Conversational Model by Applying Denoising Mechanism

MIDAS: Multi-level Intent, Domain, And Slot Knowledge Distillation for Multi-turn NLU

Transformers with Attentive Federated Aggregation for Time Series Stock Forecasting

Edge-assisted Attention-based Federated Learning for Multi-Step EVSE-enabled Prosumer Energy Demand Prediction

Federated Learning with Intermediate Representation Regularization

Query Similarity of Various Linguistic Levels for Hybridized Conversational Agents

An Explainable Artificial Intelligence Framework for Quality-Aware IoE Service Delivery

Question Difficulty Estimation with Directional Modality Association in Video Question Answering

Post-training with Data Augmentation for Improving T5-based Question Generator

A Memory Replay based Continual Learning Utilizing Class Representative and Class Boundary Data

Post-Training with Interrogative Sentences for Enhancing BART-based Korean Question Generator

CbI-M: Context-based Intervention Methodology for Rehabilitation of Persons with NCDs

WIRE: An Automated Report Generation System using Topical and Temporal Summarization

Robust Sentence Classification by Solving Out-of-Vocabulary Problem with Auxiliary Word Predictor

An End-to-End Dialog State Tracking for Task-Oriented Dialogs

An Analysis on Performance Comparison between Decoder-only and Sequence-to-Sequence Language Models in Generative NER Task

Selective Adapter Integration for Efficient Knowledge Transfer in Continual Learning

LLM-Description-Based Novel Relation Detection Framework for Open Relation Extraction

A Comparative Study on the Tokenization Strategies for Korean Standard-Dialect Translation

HTML Table Question Answering Model utilizing Special Token for Table Comprehension

Improving Schema Alignment for Closed-Domain Text-to-Data Information Extraction with Generative Language Model

Improving VQA Performance with Clarifying Questions and Resolving Ambiguous Object References

Enhancing Question Generation with LoRA-based Two-stage Training

Enhancing Performance of Cross-lingual Language Model by Advantage Actor-Critic

Transformer Adapter-Attention based Continual Learning Method for Response Generator

Low-Rank Adaptation of Large Language Model for Question Generation

Focal Loss for Sequence Generation to avoid generic answer generation

Post-training Method for Large Language Model to Enhance Question Generation

Domain Continual Learning Method without Domain Identification for Task-Oriented Dialogue Systems

Continual Learning based on Memory Replay Utilizing Class Representative and Class Boundary Data

Improving Performance of Korean Domain Question Generation Language Model through Sentence Extraction Post-Training

A Method of Retrieving High-Quality Candidate Responses for Dialogue Generation

Speech Recognition Model Ensemble Using Language Models

Triple Embedding Model for Calculating Similarity between a Sentence and a Triple

Post-Training Methods for Improving Performance of Document Summary Model

Transfer Learning-based Multi-Modal Fusion Answer Selection Model for Video Question Answering System

Continual Learning with Mimicking Human Memory System For Multi-domain Response Generator

The Study on Korean Single Document Summarization Using Pre-trained Language Models

Response-Considered Query Token Importance Weight Calculator for Dialogue Generator

Open-Domain Dialogue Model using Single-Domain Dialogue Model with Continual Learning

Hate Speech Detection Model using Noise Self-training

Sentence Compression based on Sentence Scoring Reflecting Linguistic Information

Contextual Knowledge Distillation for Task-Specific Model Compression

Model of Chatbot Answers by Language Style and Sentiment Analysis

Korean Sentence Compression Based on Unsupervised Learning Leveraging Bert

Financial Word and Label Representation Method for Financial Word Classification

Neural Question Difficulty Estimator with Bi-directional Attention in VideoQA