Publications_1 - National AI Research Lab

#Breakthrough in Neural Scaling Law

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

Year: 2025 / Author: Seil Kang, Jinyeong Kim, Junhyeok Kim, Seong Jae Hwang / Conference: International Conference on Learning Representations (ICLR)

#Breakthrough in Neural Scaling Law

WAVE: Warp-Based View Guidance for Consistent Novel View Synthesis Using a Single Image

Year: 2025 / Author: Jiwoo Park, Tae Eun Choi, Youngjun Jun, Seong Jae Hwang / Conference: International Conference on Computer Vision (ICCV)

#Breakthrough in Neural Scaling Law

Spatial Transport Optimization by Repositioning Attention Map for Training-Free Text-to-Image Synthesis

Year: 2025 / Author: Woojung Han, Yeonkyung Lee, Chanyoung Kim, Kwanghyun Park, Seong Jae Hwang / Conference: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

#Breakthrough in Neural Scaling Law

Your Large Vision-Language Model Only Needs A Few Attention Heads for Visual Grounding

Year: 2025 / Author: Seil Kang, Jinyeong Kim, Junhyeok Kim, Seong Jae Hwang / Conference: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

#Breakthrough in Neural Scaling Law

Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation

Year: 2025 / Author: Chanyoung Kim, Dayun Ju, Woojung Han, Ming-Hsuan Yang, Seong Jae Hwang / Conference: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

#Breakthrough in Neural Scaling Law

Interpreting vision transformers via residual replacement model

Year: 2025 / Author: Jinyeong Kim, Junhyeok Kim, Yumin Shim, Joohyeok Kim, Sunyoung Jung, Seong Jae Hwang / Conference: Conference on Neural Information Processing Systems (NeurIPS)

#Breakthrough in Neural Scaling Law

Rare Text Semantics Were Always There in Your Diffusion Transformer

Year: 2025 / Author: Seil Kang, Woojung Han, Dayun Ju, Seong Jae Hwang / Conference: Conference on Neural Information Processing Systems (NeurIPS)

#Breakthrough in Neural Scaling Law

Information-Theoretic Discrete Diffusion

Year: 2025 / Author: Moongyu Jeon, Sangwoo Shin, Dongjae Jeon, Albert No / Conference: Conference on Neural Information Processing Systems (NeurIPS)

#Breakthrough in Neural Scaling Law

SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment

Year: 2025 / Author: Wonje Jeung, Sangyeon Yoon, Minsuk Kahng, Albert No / Conference: Conference on Neural Information Processing Systems (NeurIPS)

#Breakthrough in Neural Scaling Law

Assigning Distinct Roles to Quantized and Low-Rank Matrices Toward Optimal Weight Decomposition

Year: 2025 / Author: Yoonjun Cho, Soeun Kim, Dongjae Jeon, Kyelim Lee, Beomsoo Lee, Albert No / Conference: Annual Meeting of the Association for Computational Linguistics (ACL)

#Breakthrough in Neural Scaling Law

R-TOFU: Unlearning in Large Reasoning Models

Year: 2025 / Author: Sangyeon Yoon, Wonje Jeung, Albert No / Conference: Conference on Empirical Methods in Natural Language Processing (EMNLP)

#Breakthrough in Neural Scaling Law

Understanding and Mitigating Memorization in Generative Models via Sharpness of Probability Landscapes

Year: 2025 / Author: Dongjae Jeon, Dueun Kim, Albert No / Conference: International Conference on Machine Learning (ICML)

#Breakthrough in Neural Scaling Law

MT-RAIG: Novel Benchmark and Evaluation Framework for Retrieval-Augmented Insight Generation over Multiple Tables

Year: 2025 / Author: Kwangwook Seo, Donguk Kwon, Dongha Lee / Conference: Annual Meeting of the Association for Computational Linguistics (ACL)

#Breakthrough in Neural Scaling Law

Unsupervised Robust Cross-Lingual Entity Alignment via Neighbor Triple Matching with Entity and Relation Texts

Year: 2025 / Author: Soojin Yoon, Sungho Ko, Tongyoung Kim, SeongKu Kang, Jinyoung Yeo, Dongha Lee / Conference: ACM International Conference on Web Search and Data Mining (WSDM)

#Breakthrough in Neural Scaling Law

Stop Playing the Guessing Game! Target-Free User Simulation for Evaluating Conversational Recommender Systems

Year: 2025 / Author: Sunghwan Kim, Kwangwook Seo, Tongyoung Kim, Jinyoung Yeo, Dongha Lee / Conference: Conference on Empirical Methods in Natural Language Processing (EMNLP)

#Breakthrough in Neural Scaling Law

Towards Personalized Conversational Sales Agents: Contextual User Profiling for Strategic Action

Year: 2025 / Author: Tongyoung Kim, Jeongeun Lee, Soojin Yoon, Sunghwan Kim, Dongha Lee / Conference: Conference on Empirical Methods in Natural Language Processing (EMNLP)

#Breakthrough in Neural Scaling Law

Can Large Language Models be Effective Online Opinion Miners?

Year: 2025 / Author: Ryang Heo, Yongsik Seo, Junseong Lee, Dongha Lee / Conference: Conference on Empirical Methods in Natural Language Processing (EMNLP)

#Breakthrough in Neural Scaling Law

Representing 3D Shapes With 64 Latent Vectors for 3D Diffusion Models

Year: 2025 / Author: In Cho, Youngbeom Yoo, Subin Jeon, Seon Joo Kim / Conference: International Conference on Computer Vision (ICCV)

#Breakthrough in Neural Scaling Law

Robust and Consistent Online Video Instance Segmentation via Instance Mask Propagation

Year: 2025 / Author: Miran Heo, Seoung Wug Oh, Seon Joo Kim, Joon-Young Lee / Conference: The Association for the Advancement of Artificial Intelligence (AAAI)

#Breakthrough in Neural Scaling Law

AMQ: Enabling AutoML for Mixed-precision Weight-Only Quantization of Large Language Models

Year: 2025 / Author: Sangjun Lee, Seung-taek Woo, Jungyu Jin, Changhun Lee, Eunhyeok Park / Conference: Conference on Empirical Methods in Natural Language Processing (EMNLP)

Publications