7 citations · 48 across the 69 of their papers we have counts for
20 papers · 1 filter
Targeted Data Protection for Diffusion Model by Matching Training Trajectory
Hojun Lee, Mijin Koo, Yeji Song +1
Recent advancements in diffusion models have made fine-tuning text-to-image models for personalization increasingly accessible, but have also raised significant concerns regarding…
Multi-dimensional Preference Alignment by Conditioning Reward Itself
Jiho Jang, Jinyoung Kim, Kyungjune Baek +1
Reinforcement Learning from Human Feedback has emerged as a standard for aligning diffusion models. However, we identify a fundamental limitation in the standard DPO formulation be…
LoGoColor: Local-Global 3D Colorization for 360° Scenes
Yeonjin Chang, Juhwan Cho, Seunghyeon Seo +2
Single-channel 3D reconstruction is widely used in fields such as robotics and medical imaging. While these methods are good at reconstructing 3D geometry, their outputs are typica…
Point-to-Point: Sparse Motion Guidance for Controllable Video Editing
Yeji Song, Jaehyun Lee, Mijin Koo +2
Accurately preserving motion while editing a subject remains a core challenge in video editing tasks. Existing methods often face a trade-off between edit and motion fidelity, as t…
Safety-Preserving PTQ via Contrastive Alignment Loss
Sunghyun Wee, Suyoung Kim, Hyeonjin Kim +2
Post-Training Quantization (PTQ) has become the de-facto standard for efficient LLM deployment, yet its optimization objective remains fundamentally incomplete. Standard PTQ method…
Uncovering the Potential Risks in Unlearning: Danger of English-only Unlearning in Multilingual LLMs
Kyomin Hwang, Hyeonjin Kim, Seungyeon Kim +2
There have been a couple of studies showing that attempting to erase multilingual knowledge using only English data is insufficient for multilingual LLMs. However, their analyses r…
Unlocking the Potential of Diffusion Language Models through Template Infilling
Junhoo Lee, Seungyeon Kim, Nojun Kwak
Diffusion Language Models (DLMs) have emerged as a promising alternative to Autoregressive Language Models, yet their inference strategies remain limited to prefix-based prompting…
Deep Edge Filter: Return of the Human-Crafted Layer in Deep Learning
Dongkwan Lee, Junhoo Lee, Nojun Kwak
We introduce the Deep Edge Filter, a novel approach that applies high-pass filtering to deep neural network features to improve model generalizability. Our method is motivated by o…
Causal Interpretation of Sparse Autoencoder Features in Vision
Sangyu Han, Yearim Kim, Nojun Kwak
Understanding what sparse auto-encoder (SAE) features in vision transformers truly represent is usually done by inspecting the patches where a feature's activation is highest. Howe…
The Role of Teacher Calibration in Knowledge Distillation
Suyoung Kim, Seonguk Park, Junhoo Lee +1
Knowledge Distillation (KD) has emerged as an effective model compression technique in deep learning, enabling the transfer of knowledge from a large teacher model to a compact stu…
Style Composition within Distinct LoRA modules for Traditional Art
Jaehyun Lee, Wonhark Park, Wonsik Shin +3
Diffusion-based text-to-image models have achieved remarkable results in synthesizing diverse images from text prompts and can capture specific artistic styles via style personaliz…
What's Making That Sound Right Now? Video-centric Audio-Visual Localization
Hahyeon Choi, Junhoo Lee, Nojun Kwak
Audio-Visual Localization (AVL) aims to identify sound-emitting sources within a visual scene. However, existing studies focus on image-level audio-visual associations, failing to…
ReFlex: Text-Guided Editing of Real Images in Rectified Flow via Mid-Step Feature Extraction and Attention Adaptation
Jimyeong Kim, Jungwon Park, Yeji Song +2
Rectified Flow text-to-image models surpass diffusion models in image quality and text alignment, but adapting ReFlow for real-image editing remains challenging. We propose a new r…
Generative Head-Mounted Camera Captures for Photorealistic Avatars
Shaojie Bai, Seunghyeon Seo, Yida Wang +8
Enabling photorealistic avatar animations in virtual and augmented reality (VR/AR) has been challenging because of the difficulty of obtaining ground truth state of faces. It is ph…
S3D: Sketch-Driven 3D Model Generation
Hail Song, Wonsik Shin, Naeun Lee +3
Generating high-quality 3D models from 2D sketches is a challenging task due to the inherent ambiguity and sparsity of sketch data. In this paper, we present S3D, a novel framework…
Understanding Differential Transformer Unchains Pretrained Self-Attentions
Chaerin Kong, Jiho Jang, Nojun Kwak
Differential Transformer has recently gained significant attention for its impressive empirical performance, often attributed to its ability to perform noise canceled attention. Ho…
Unlocking the Potential of Unlabeled Data in Semi-Supervised Domain Generalization
Dongkwan Lee, Kyomin Hwang, Nojun Kwak
We address the problem of semi-supervised domain generalization (SSDG), where the distributions of train and test data differ, and only a small amount of labeled data along with a…
A Revisit to the Decoder for Camouflaged Object Detection
Seung Woo Ko, Joopyo Hong, Suyoung Kim +5
Camouflaged object detection (COD) aims to generate a fine-grained segmentation map of camouflaged objects hidden in their background. Due to the hidden nature of camouflaged objec…
DivCon-NeRF: Diverse and Consistent Ray Augmentation for Few-Shot NeRF
Ingyun Lee, Jae Won Jang, Seunghyeon Seo +1
Neural Radiance Field (NeRF) has shown remarkable performance in novel view synthesis but requires numerous multi-view images, limiting its practicality in few-shot scenarios. Ray…
ROODI: Reconstructing Occluded Objects with Denoising Inpainters
Yeonjin Chang, Erqun Dong, Seunghyeon Seo +2
While the quality of novel-view images has improved dramatically with 3D Gaussian Splatting, extracting specific objects from scenes remains challenging. Isolating individual 3D Ga…