38 citations · 111 across the 20 of their papers we have counts for
21 papers
MaDeL: Manifold-Decomposed Feature Losses for Generative Modeling
Beomsu Kim, Jong Chul Ye, Kwanyoung Kim
Generative models are often trained with isotropic objectives such as mean-squared error. For data concentrated near a low-dimensional manifold, however, such losses conflate displ…
Are Visual Place Recognition Models Recognizing Places or Conditions? Distractor-Augmented Evaluation and Condition Suppression
Beomsu Kim, Minwoo Jung, Giseop Kim
Long-term Visual Place Recognition (VPR) is typically evaluated by matching queries from one condition against a database from another. Crowdsourced map databases, however, may mix…
Align Your Tangent: Training Better Consistency Models via Manifold-Aligned Tangents
Beomsu Kim, Byunghee Cha, Jong Chul Ye
With diffusion and flow matching models achieving state-of-the-art generating performance, the interest of the community now turned to reducing the inference time without sacrifici…
Boost-and-Skip: A Simple Guidance-Free Diffusion for Minority Generation
Soobin Um, Beomsu Kim, Jong Chul Ye
Minority samples are underrepresented instances located in low-density regions of a data manifold, and are valuable in many generative AI applications, such as data augmentation, c…
Latent Schrodinger Bridge: Prompting Latent Diffusion for Fast Unpaired Image-to-Image Translation
Jeongsol Kim, Beomsu Kim, Jong Chul Ye
Diffusion models (DMs), which enable both image generation from noise and inversion from data, have inspired powerful unpaired image-to-image (I2I) translation algorithms. However,…
Simple ReFlow: Improved Techniques for Fast Flow Models
Beomsu Kim, Yu-Guan Hsieh, Michal Klein +4
Diffusion and flow-matching models achieve remarkable generative performance but at the cost of many sampling steps, this slows inference and limits applicability to time-critical…
(Deep) Generative Geodesics
Beomsu Kim, Michael Puthawala, Jong Chul Ye +1
In this work, we propose to study the global geometrical properties of generative models. We introduce a new Riemannian metric to assess the similarity between any two data points.…
Generalized Consistency Trajectory Models for Image Manipulation
Beomsu Kim, Jaemin Kim, Jeongsol Kim +1
Diffusion models (DMs) excel in unconditional generation, as well as on applications such as image editing and restoration. The success of DMs lies in the iterative nature of diffu…
CEmb-SAM: Segment Anything Model with Condition Embedding for Joint Learning from Heterogeneous Datasets
Dongik Shin, Beomsuk Kim, Seungjun Baek
Automated segmentation of ultrasound images can assist medical experts with diagnostic and therapeutic procedures. Although using the common modality of ultrasound, one typically n…
Energy-Based Cross Attention for Bayesian Context Update in Text-to-Image Diffusion Models
Geon Yeong Park, Jeongsol Kim, Beomsu Kim +2
Despite the remarkable performance of text-to-image diffusion models in image generation tasks, recent studies have raised the issue that generated images sometimes cannot capture…
Unpaired Image-to-Image Translation via Neural Schrödinger Bridge
Beomsu Kim, Gihyun Kwon, Kwanyoung Kim +1
Diffusion models are a powerful class of generative models which simulate stochastic differential equations (SDEs) to generate data from noise. While diffusion models have achieved…
Minimizing Trajectory Curvature of ODE-based Generative Models
Sangyun Lee, Beomsu Kim, Jong Chul Ye
Recent ODE/SDE-based generative models, such as diffusion models, rectified flows, and flow matching, define a generative process as a time reversal of a fixed forward process. Eve…
Measuring and Improving Semantic Diversity of Dialogue Generation
Seungju Han, Beomsu Kim, Buru Chang
Response diversity has become an important criterion for evaluating the quality of open-domain dialogue generation models. However, current evaluation metrics for response diversit…
Denoising MCMC for Accelerating Diffusion-Based Generative Models
Beomsu Kim, Jong Chul Ye
Diffusion models are powerful generative models that simulate the reverse of diffusion processes using score functions to synthesize data from noise. The sampling process of diffus…
Mitigating Out-of-Distribution Data Density Overestimation in Energy-Based Models
Beomsu Kim, Jong Chul Ye
Deep energy-based models (EBMs), which use deep neural networks (DNNs) as energy functions, are receiving increasing attention due to their ability to learn complex distributions.…
Meet Your Favorite Character: Open-domain Chatbot Mimicking Fictional Characters with only a Few Utterances
Seungju Han, Beomsu Kim, Jin Yong Yoo +4
In this paper, we consider mimicking fictional characters as a promising direction for building engaging conversation models. To this end, we present a new practical task where onl…
Energy-Based Contrastive Learning of Visual Representations
Beomsu Kim, Jong Chul Ye
Contrastive learning is a method of learning visual representations by training Deep Neural Networks (DNNs) to increase the similarity between representations of positive pairs (tr…
Understanding and Improving the Exemplar-based Generation for Open-domain Conversation
Seungju Han, Beomsu Kim, Seokjun Seo +2
Exemplar-based generative models for open-domain conversation produce responses based on the exemplars provided by the retriever, taking advantage of generative models and retrieva…
Disentangling Label Distribution for Long-tailed Visual Recognition
Youngkyu Hong, Seungju Han, Kwanghee Choi +3
The current evaluation protocol of long-tailed visual recognition trains the classification model on the long-tailed source label distribution and evaluates its performance on the…
MarioNETte: Few-shot Face Reenactment Preserving Identity of Unseen Targets
Sungjoo Ha, Martin Kersner, Beomsu Kim +2
When there is a mismatch between the target identity and the driver identity, face reenactment suffers severe degradation in the quality of the result, especially in a few-shot set…
Temporal Convolution for Real-time Keyword Spotting on Mobile Devices
Seungwoo Choi, Seokjun Seo, Beomjun Shin +5
Keyword spotting (KWS) plays a critical role in enabling speech-based user interactions on smart devices. Recent developments in the field of deep learning have led to wide adoptio…