arXiv subjects

Wooseok Kim

Publications and source records attributed to Wooseok Kim.

4 recordsLinked to original sources

arXiv2026

Voltron: Enabling Elastic Multi-Device Execution of LLM Inference for Empowered Edge Intelligence

Large language models (LLMs) are widely used in intelligent services due to their remarkable capability in generative tasks. Typically, LLM-based services process the inference requests of the users in a centralized data center. Unfortunately, such centralized execution has limitations for end-users, such as increased response latency with communication overhead and privacy leakage risk. To alleviate the aforementioned limitations, there have been increasing pushes to execute LLM inference locally on user-end devices. However, the limited resources of a single edge device impose restrictions on achievable accuracy of LLMs. To overcome the issue, we first propose to leverage multiple user-end devices available at the edge for LLM inference, enabling the execution of larger models. Specifically, we propose Voltron, a novel on-device LLM inference framework that elastically utilizes multiple user-end devices for LLM inference execution while adapting to diverse real-world edge environments. In our evaluation, Voltron achieves up to 16.5% higher accuracy than state-of-the-art LLMs that can be executed on a single edge device, satisfying user QoS requirements.

cs.DC↗

arXiv2026

Deep Filter Estimation from Inter-Frame Correlations for Monaural Speech Dereverberation

Speech dereverberation in distant-microphone scenarios remains challenging due to the high correlation between reverberation and target signals, often leading to poor generalization in real-world environments. We propose IF-CorrNet, a correlation-to-filter architecture designed for robustness against acoustic variability. Unlike conventional black-box mapping methods that directly estimate complex spectra, IF-CorrNet explicitly exploits inter-frame STFT correlations to estimate multi-frame deep filters for each time-frequency bin. By shifting the learning objective from direct mapping to filter estimation, the network effectively constrains the solution space, which simplifies the training process and mitigates overfitting to synthetic data. Experimental results on the REVERB Challenge dataset demonstrate that IF-CorrNet achieves a substantial gain in the SRMR metric on RealData, confirming its robustness in suppressing reverberation and noise in practical, non-synthetic environments.

eess.AS↗

arXiv2024

Point-Cloud Based Inverse Design of Free-Form Metamaterials Using Deep Generative Networks

Mechanical metamaterials enable precise control over structural properties, but their design method remains challenging due to their complex structure. Although additive manufacturing has expanded geometric freedom, navigating this vast and complex design space still requires computationally intensive simulations or expert-driven processes. Recently, artificial intelligence (AI)-driven design approaches have emerged to address these limitations, but many studies restrict their scope to parametric representations, limiting their generative capacity to predefined shapes. Here, we present a point cloud-based generative framework that enables the inverse design of 3D metamaterial without parametric constraints. Trained on a number of structurally valid unit cells, the present machine learning model learns geometric patterns, mitigates common connectivity issues inherent in point cloud generation. The proposed model constructs a latent space organized by mechanical properties and naturally clustered by unit cell types. By sampling this latent space, our method supports both property-guided inverse design and generation of topologically gradient transition between distinct unit cell types. This approach facilitates inverse design of 3D metamaterials with high geometric complexity.

cond-mat.soft↗

arXiv2024

REF$^2$-NeRF: Reflection and Refraction aware Neural Radiance Field

Recently, significant progress has been made in the study of methods for 3D reconstruction from multiple images using implicit neural representations, exemplified by the neural radiance field (NeRF) method. Such methods, which are based on volume rendering, can model various light phenomena, and various extended methods have been proposed to accommodate different scenes and situations. However, when handling scenes with multiple glass objects, e.g., objects in a glass showcase, modeling the target scene accurately has been challenging due to the presence of multiple reflection and refraction effects. Thus, this paper proposes a NeRF-based modeling method for scenes containing a glass case. In the proposed method, refraction and reflection are modeled using elements that are dependent and independent of the viewer's perspective. This approach allows us to estimate the surfaces where refraction occurs, i.e., glass surfaces, and enables the separation and modeling of both direct and reflected light components. The proposed method requires predetermined camera poses, but accurately estimating these poses in scenes with glass objects is difficult. Therefore, we used a robotic arm with an attached camera to acquire images with known poses. Compared to existing methods, the proposed method enables more accurate modeling of both glass refraction and the overall scene.

cs.CV↗