Publications_1 - National AI Research Lab

#Breakthrough in Neural Scaling Law

Retrieval-Augmented Generation with Estimation of Source Reliability

연도: 2025 / 저자: 황정연, 박준영, 박혜진, 김동우, 박상돈, 옥정슬 / 학술대회명: Conference on Empirical Methods in Natural Language Processing (EMNLP)

#Breakthrough in Neural Scaling Law

Adaptive Non-uniform Timestep Sampling for Accelerating Diffusion Model Training

연도: 2025 / 저자: 김면수, 기동현, 심성웅, 이병준 / 학술대회명: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

#Breakthrough in Neural Scaling Law

K/DA: Automated Data Generation Pipeline for Detoxifying Implicitly Offensive Language in Korean

연도: 2025 / 저자: 전민경, 정혜민, 김예랑, 김지영, 조재현, 이병준 / 학술대회명: Annual Meeting of the Association for Computational Linguistics (ACL)

#Breakthrough in Neural Scaling Law

Iterative Prompt Refinement for Safer Text-to-Image Generation

연도: 2025 / 저자: 전진우, 오준혁, 이하영, 이병준 / 학술대회명: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP)

#Breakthrough in Neural Scaling Law

Rethinking DPO: The Role of Rejected Responses in Preference Misalignment

연도: 2025 / 저자: 조재현, 오준혁, 김면수, 이병준 / 학술대회명: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP)

#Breakthrough in Neural Scaling Law

NBDI: A Simple and Effective Termination Condition for Skill Extraction from Task-Agnostic Demonstrations

연도: 2025 / 저자: 김면수, 이하영, 심성웅, 서준호, 이병준 / 학술대회명: International Conference on Machine Learning (ICML)

#Breakthrough in Neural Scaling Law

Towards Generalizable 3D Human Pose Estimation via Ensembles on Flat Loss Landscapes

연도: 2025 / 저자: 한주민, 김준희, 이성환 / 학술대회명: Conference on Neural Information Processing Systems (NeurIPS)

#Breakthrough in Neural Scaling Law

GRAIL: Gradient-Based Adaptive Unlearning for Privacy and Copyright in LLMs

연도: 2025 / 저자: 김건우, 박지훈, 한주민, 이성환 / 학술대회명: International Joint Conference on Neural Networks (IJCNN)

#Breakthrough in Neural Scaling Law

Towards Fine-Grained Interpretability: Counterfactual Explanations for Misclassification with Saliency Partition

연도: 2025 / 저자: Lintong Zhang, Kang Yin, 이성환 / 학술대회명: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

#Breakthrough in Neural Scaling Law

DiGIT: Multi-Dilated Gated Encoder and Central-Adjacent Region Integrated Decoder for Temporal Action Detection Transformer

연도: 2025 / 저자: 김호중, 이예랑, 홍정호, 이성환 / 학술대회명: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

#Breakthrough in Neural Scaling Law

Integrating Locality-Aware Attention with Transformers for General Geometry PDEs

연도: 2025 / 저자: 고민수, 박범철, 공희조, 이성환 / 학술대회명: International Joint Conference on Neural Networks (IJCNN)

#Breakthrough in Neural Scaling Law

ACoRN: Noise-Robust Abstractive Compression in Retrieval-Augmented Language Models

연도: 2025 / 저자: 김신곤, 정건호, 이성환 / 학술대회명: International Joint Conference on Neural Networks (IJCNN)

#Breakthrough in Neural Scaling Law

Confidence-Weighted Boundary-Aware Learning for Semi-Supervised Semantic Segmentation

연도: 2025 / 저자: Ebenezer Tarubinga, Jenifer Kalafatovich, 이성환 / 학술대회명: International Joint Conference on Neural Networks (IJCNN)

#Breakthrough in Neural Scaling Law

ProPose: Probabilistic 3D Human Pose Estimation with Instance-Level Distribution and Normalizing Flow

연도: 2025 / 저자: 한주민, 김준희, 이성환 / 학술대회명: Association for the Advancement of Artificial Intelligence (AAAI)

#Breakthrough in Neural Scaling Law

FairDICE: Fairness-Driven Offline Multi-Objective Reinforcement Learning

연도: 2025 / 저자: 김우성, 이진호, 이종민, 이병준 / 학술대회명: Advances in Neural Information Processing Systems (NeurIPS)

#Breakthrough in Neural Scaling Law

PLATYPUS: Progressive Local Surface Estimator for Arbitrary-Scale Point Cloud Upsampling

연도: 2025 / 저자: 김동현, 권현경, 김유민, 황성재 / 학술대회명: The Association for the Advancement of Artificial Intelligence (AAAI)

#Breakthrough in Neural Scaling Law

Mono-Modalizing Extremely Heterogeneous Multi-Modal Medical Image Registration

연도: 2025 / 저자: 추교빈, 한현경, 김진영, 윤찬용, 황성재 / 학술대회명: Medical Image Computing and Computer Assisted Intervention (MICCAI)

#Breakthrough in Neural Scaling Law

Pathology-Aware Adaptive Watermarking for Text-Driven Medical Image Synthesis

연도: 2025 / 저자: 김찬영, 주다윤, 김진영, 한우정, Roberto Alcover-Couso, 황성재 / 학술대회명: Medical Image Computing and Computer Assisted Intervention (MICCAI)

#Breakthrough in Neural Scaling Law

PRETI: Patient-Aware Retinal Foundation Model via Metadata-Guided Representation Learning

연도: 2025 / 저자: 이연경, 한우정, 전영준, 김현민, 조정경, 황성재 / 학술대회명: Medical Image Computing and Computer Assisted Intervention (MICCAI)

#Breakthrough in Neural Scaling Law

See What You Are Told: Visual Attention Sink in Large Multimodal Models

연도: 2025 / 저자: 강세일, 김진영, 김준혁, 황성재 / 학술대회명: International Conference on Learning Representations (ICLR)

연구논문