Jisu Kim

Changhyeon Kim

Kang Min Lee

Deokyeong Kang

Jii Cha

Yejin Yoon

Sung Hoon Lim

Taejun Yun

Young Hyun Yoo

Jinhyeon Kim

Welcome to Natural Language Processing (NLP) Lab. at Hanyang University. We study various approaches and problems with regard to natural language, chiefly based on machine learning and AI technologies. We are looking for MS/Ph.D. students (and interns) who are self-motivated and passionate about doing research in NLP. Our research includes develop AI translation technology grounded in the Korean cultural context (한국문화 맥락 기반 인공지능 번역 기술 개발). We participate in the consortium led by Naver Cloud to conduct research and development for the National AI Foundation Model Project (독자 AI 파운데이션 모델 프로젝트). We receive a three-year grant from the NRF for the project titled "Conversational Agents with Hyper Long-Term Memory" (Outstanding Young Scientist Grants).

한양대학교 자연어 처리(NLP) 연구실에 오신 것을 환영합니다. 우리는 주로 기계 학습 및 AI 기술을 기반으로 자연어와 관련된 다양한 접근 방식과 문제를 연구합니다. 우리는 NLP 연구에 대한 동기 부여와 열정을 가진 석/박사 학생(및 인턴)을 찾고 있습니다. 우리의 연구 프로젝트에는 한국 문화 맥락 기반 인공지능 번역 기술 개발(한국콘텐츠진흥원), 네이버 클라우드가 주도하는 컨소시엄에 참여하여 독자 AI 파운데이션 모델 프로젝트 연구 개발 수행, 그리고 "초장기 기억을 가진 대화형 에이전트" 프로젝트로 NRF로부터 3년 지원금 수령(우수 젊은 과학자 연구비)이 포함됩니다.

ADVICE: Answer-Dependent Verbalized Confidence Estimation

OMHBench: Benchmarking Balanced and Grounded Omni-Modal Multi-Hop Reasoning

Beyond Task-Oriented and Chitchat Dialogues: Proactive and Transition-Aware Conversational Agents

MAGIC: A Multi-Hop and Graph-Based Benchmark for Inter-Context Conflicts in Retrieval-Augmented Generation

Memorization or Reasoning? Exploring the Idiom Understanding of LLMs

Does Localization Inform Unlearning? A Rigorous Examination of Local Parameter Attribution for Knowledge Unlearning in Language Models

FCMR: Robust Evaluation of Financial Cross-Modal Multi-Hop Reasoning

Latent Preference Modeling for Cross-Session Personalized Tool Calling

OmniAC Bench: A Benchmark for Evaluating Context-Grounded Acoustic Control in Omni-Modal Models

UniKnow: A Unified Framework for Reliable Language Model Behavior across Parametric and External Knowledge

Investigating the Influence of Prompt-Specific Shortcuts in AI Generated Text Detection

Query Lens: Interpreting Sparse Key-Value Features with Indirect Effects

Uncertainty-Aware Multi-Metric Evaluation of Human–Machine Agreement for LLM-Based Educational Assessment

Superficial Success vs. Internal Breakdown: An Empirical Study of Generalization in Adaptive Multi-Agent Systems

Hybrid-Vector Retrieval for Visually Rich Documents: Combining Single-Vector Efficiency and Multi-Vector Accuracy

Think Just Enough: Leveraging Self-Assessed Confidence for Adaptive Reasoning in Language Models

Development and Evaluation of a Dual-Expertise, Utterance-Level Framework for LLM-Based Science Classroom Discourse Analysis

When to Speak, When to Abstain: Contrastive Decoding with Abstention

ENGinius: A Bilingual LLM Optimized for Plant Construction Engineering

KGMEL: Knowledge Graph-Enhanced Multimodal Entity Linking

Subgraph-Aware Training of Language Models for Knowledge Graph Completion Using Structure-Aware Contrastive Learning

ESPRESSO: An Effective Approach to Passage Retrieval for High-Quality Conversational Recommender Systems

Adaptive Contrastive Decoding in Retrieval-Augmented Generation for Handling Noisy Contexts

Revisiting the Impact of Pursuing Modularity for Code Generation

Aligning Language Models to Explicitly Handle Ambiguity

Hyper-CL: Conditioning Sentence Representations with Hypernetworks

Analysis of Multi-Source Language Training in Cross-Lingual Transfer

BlendX: Complex Multi-Intent Detection with Blended Patterns

X-SNS: Cross-Lingual Transfer Prediction through Sub-Network Similarity

Universal Domain Adaptation for Robust Handling of Distributional Shifts in NLP

Prompt-Augmented Linear Probing: Scaling Beyond The Limit of Few-shot In-Context Learners

Ground-Truth Labels Matter: A Deeper Look into Input-Label Demonstrations

Enhancing Out-of-Distribution Detection in Natural Language Understanding via Implicit Layer Ensemble

Revisiting the Practical Effectiveness of Constituency Parse Extraction from Pre-trained Language Models

Multilingual Chart-based Constituency Parse Extraction from Pre-trained Language Models

Self-Guided Contrastive Learning for BERT Sentence Representations

Heads-up! Unsupervised Constituency Parsing via Self-Attention Heads

Are Pre-trained Language Models Aware of Phrases? Simple but Strong Baselines for Grammar Induction

Cell-aware Stacked LSTMs for Modeling Sentences

Don't Just Scratch the Surface: Enhancing Word Representations for Korean with Hanja

Intrinsic Evaluation of Grammatical Information within Word Embeddings

A Cross-Sentence Latent Variable Model for Semi-Supervised Text Sequence Matching

Dynamic Compositionality in Recursive Neural Networks with Structure-aware Tag Representations

Element-wise Bilinear Interaction for Sentence Matching

거대 언어 모델 관용구 처리의 역전의 저주 현상 탐구 (Reversal Curse in Idiomatic Tasks of Large Language Models)

대화 맥락 추론 향상을 위한 리즈닝 피드백 기반 학습 (Enhancing Conversational Context Inference via Reasoning Feedback-Based Learning)

한국어 표 설명 능력 향상을 위한 전처리 및 학습 방법론 탐구 (Learning Strategies to Improve Table Understanding and Explanation in Korean)

KR-HumanEval을 활용한 언어 모델의 한국어 프로그램 합성 성능 분석 (Analysis of Language Models in Korean Program Synthesis Based on the KR-HumanEval Benchmark)

한국어 발화의 다중 의도 감지 연구 (Multi-Intent Detection for Korean Spoken Language)

다중 레이블 분류를 위한 프롬프팅 고도화 (Enhanced Prompting for Multi-Label Classification)

생락과 상호참조를 보강한 다중 의도 데이터셋 (A Multi-Intent Dataset Enhanced with Implicit Concatenation)

기계 독해를 활용한 한국어 의미역 결정 (Korean Semantic Role Labeling with Machine Reading Comprehension)

대조학습 기반 문장표현 방법론 개선을 위한 공통 오류 분석 및 앙상블 기법 (Enhancing Sentence Representations with Common Error Analysis and Ensemble Techniques in Contrastive Learning)

효과적인 한국어 교차언어 전송을 위한 특성 연구 (Research on Features for Effective Cross-Lingual Transfer in Korean)

MAdapter: 효율적인 중간 층 도입을 통한 Adapter 구조 개선 (MAdapter: A Refinement of Adapters by Augmenting Efficient Middle Layers)

원천 언어 다각화를 통한 교차 언어 전이 성능 향상 (Enhanced Zero-Shot Cross-Lingual Transfer with the Diversification of Source Languages)

한국어 문장 표현을 위한 비지도 대조 학습 방법론의 비교 및 분석 (Comparison and Analysis of Unsupervised Contrastive Learning Approaches for Korean Sentence Representations)

RAISE: Enhancing Scientific Reasoning in LLMs via Step-by-Step Retrieval

Self-Generated In-Context Learning: Leveraging Auto-regressive Language Models as a Demonstration Generator

HYU at SemEval-2022 Task 2: Effective Idiomaticity Detection with Consideration at Different Levels of Contextualization

IDS at SemEval-2020 Task 10: Does Pre-trained Language Model Know What to Emphasize?

Summary Level Training of Sentence Rewriting for Abstractive Summarization

SNU_IDS at SemEval-2018 Task 12: Sentence Encoder with Contextualized Vectors for Argument Reasoning Comprehension

A Syllable-based Technique for Word Embeddings of Korean Words

Administrative Decision-Making with Generative AI: The Challenge of Epistemic Boundedness

대규모 언어 모델의 지식 삭제 연구 동향 (A Survey on Machine Unlearning for Large Language Models)

한국어 의미역 결정 모형을 통한 텍스트 내 서사 구조 분석 방법 (A Framework for Narrative Structure Analysis Using Automated Semantic Role Labeling in Korean Text)

Comparison and Analysis of Unsupervised Contrastive Learning Approaches for Korean Sentence Representations

HYU Natural Language Processing (NLP) Laboratory

교수님 소개

김태욱

연구실 소개

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Latent Preference Modeling for Cross-Session Personalized Tool Calling

OmniAC Bench: A Benchmark for Evaluating Context-Grounded Acoustic Control in Omni-Modal Models

Query Lens: Interpreting Sparse Key-Value Features with Indirect Effects

연구실 보유 특허

대학별 입시끈

우리 연구실 모집

교수님 소개

김태욱

연구실 키워드

알럼나이가 간 회사

연구실 지원 방법

대우 조건

연도별 논문 인용수

저널 & 학회 논문 출판 수

연구실 논문 목록

Latent Preference Modeling for Cross-Session Personalized Tool Calling

OmniAC Bench: A Benchmark for Evaluating Context-Grounded Acoustic Control in Omni-Modal Models

Query Lens: Interpreting Sparse Key-Value Features with Indirect Effects

연구실 보유 특허

최근 3년간 연구실적 높은 연구실

대학별 입시끈

우리 연구실 모집