arXiv subjects

Youngmin Kim

Publications and source records attributed to Youngmin Kim.

At least 37 records · Page 2Linked to original sources

arXiv2025

Scalp Diagnostic System With Label-Free Segmentation and Training-Free Image Translation

Scalp disorders are highly prevalent worldwide, yet remain underdiagnosed due to limited access to expert evaluation and the high cost of annotation. Although AI-based approaches hold great promise, their practical deployment is hindered by challenges such as severe data imbalance and the absence of pixel-level segmentation labels. To address these issues, we propose ScalpVision, an AI-driven system for the holistic diagnosis of scalp diseases. In ScalpVision, effective hair segmentation is achieved using pseudo image-label pairs and an innovative prompting method in the absence of traditional hair masking labels. Additionally, ScalpVision introduces DiffuseIT-M, a generative model adopted for dataset augmentation while maintaining hair information, facilitating improved predictions of scalp disease severity. Our experimental results affirm ScalpVision's efficiency in diagnosing a variety of scalp conditions, showcasing its potential as a valuable tool in dermatological care. Our code is available at https://github.com/winston1214/ScalpVision.

cs.CV↗

arXiv2025

A 2-Stage Model for Vehicle Class and Orientation Detection with Photo-Realistic Image Generation

We aim to detect the class and orientation of a vehicle by training a model with synthetic data. However, the distribution of the classes in the training data is imbalanced, and the model trained on the synthetic image is difficult to predict in real-world images. We propose a two-stage detection model with photo-realistic image generation to tackle this issue. Our model mainly takes four steps to detect the class and orientation of the vehicle. (1) It builds a table containing the image, class, and location information of objects in the image, (2) transforms the synthetic images into real-world images style, and merges them into the meta table. (3) Classify vehicle class and orientation using images from the meta-table. (4) Finally, the vehicle class and orientation are detected by combining the pre-extracted location information and the predicted classes. We achieved 4th place in IEEE BigData Challenge 2022 Vehicle class and Orientation Detection (VOD) with our approach.

cs.CV↗

arXiv2025

Speaking Beyond Language: A Large-Scale Multimodal Dataset for Learning Nonverbal Cues from Video-Grounded Dialogues

Nonverbal communication is integral to human interaction, with gestures, facial expressions, and body language conveying critical aspects of intent and emotion. However, existing large language models (LLMs) fail to effectively incorporate these nonverbal elements, limiting their capacity to create fully immersive conversational experiences. We introduce MARS, a multimodal language model designed to understand and generate nonverbal cues alongside text, bridging this gap in conversational AI. Our key innovation is VENUS, a large-scale dataset comprising annotated videos with time-aligned text, facial expressions, and body language. Leveraging VENUS, we train MARS with a next-token prediction objective, combining text with vector-quantized nonverbal representations to achieve multimodal understanding and generation within a unified framework. Based on various analyses of the VENUS datasets, we validate its substantial scale and high effectiveness. Our quantitative and qualitative results demonstrate that MARS successfully generates text and nonverbal languages, corresponding to conversational input.

cs.AI↗

arXiv2025

Mantra: Rewriting Quantum Programs to Minimize Trap-Movements for Zoned Rydberg Atom Arrays

A zoned neutral atom architecture achieves exceptional fidelity by segregating the execution spaces of 1- and 2-qubit gates, being a promising candidate for high-accuracy quantum systems. Unfortunately, naively applying programs designed for static qubit topologies to zoned architectures may result in most execution time being consumed by inter-zone travels of atoms. To address this, we introduce Mantra (Minimizing trAp movemeNts for aTom aRray Architectures), which rewrites quantum programs to reduce the interleaving of single- and two-qubit gates. Mantra incorporates three strategies: (i) a fountain-shaped controlled-Z (CZ) chain, (ii) ZZ-interaction protocol without a 1-qubit gate, and (iii) preemptive gate scheduling. Mantra reduces inter-zone movements by 68%, physical gate counts by 35%, and improves circuit fidelities by 17% compared to the standard executions.

quant-ph↗

arXiv2025

Dynamic Tuning of Single-Photon Emission in Monolayer WSe2 via Localized Strain Engineering

Two-dimensional (2D) materials have emerged as promising candidates for next-generation integrated single-photon emitters (SPEs). However, significant variability in the emission energies of 2D SPEs presents a major challenge in producing identical single photons from different SPEs, which may become crucial for various quantum applications including quantum information processing. Although various approaches to dynamically tuning the emission energies of 2D SPEs have been developed to address the issue, the practical solution to matching multiple individual SPEs in a single 2D flake is still scarce. In this work, we demonstrate a precise emission energy tuning of individual SPEs in a WSe2 monolayer. Our approach utilizes localized strain fields near individual SPEs, which we control independently by adjusting the physical volume of an SU-8-based stressor layer via focused laser annealing. This technique allows continuous emission energy tuning of up to 15 meV while maintaining the qualities of SPEs. Additionally, we showcase the precise spectral alignment of three distinct SPEs in a single WSe2 monolayer to the same wavelength. The tunability of 2D SPEs represents a solid step towards the on-chip integrated photonics with 2D materials for quantum technologies.

cond-mat.mes-hall↗

arXiv2025

Balancing Thermal Relaxation Deviations of Near-Future Quantum Computing Results via Bit-Inverted Programs

One of the predominant causes of program distortion in the real quantum computing system may be attributed to the probability deviation caused by thermal relaxation. We introduce Barber (Balancing reAdout Results using Bit-invErted ciRcuits), a method designed to counteract the asymmetric thermal relaxation deviation and improve the reliability of near-term quantum programs. Barber collaborates with a bit-inverted quantum circuit, where the excited quantum state of qubits is assigned to the $\lvert 0 \rangle$ and the unexcited state to the $\lvert 1 \rangle$. In doing so, bit-inverted quantum circuits can experience thermal relaxation in the opposite direction compared to standard quantum circuits. Barber can effectively suppress the thermal relaxation deviation in program's readout results by selectively merging distributions from the standard and bit-inverted circuits.

quant-ph↗

arXiv2025

Mid-infrared group-IV nanowire laser

Semiconductor nanowires have shown great potential for enabling ultra-compact lasers for integrated photonics platforms. Despite the impressive progress in developing nanowire lasers, their integration into Si photonics platforms remains challenging largely due to the use of III-V and II-VI semiconductors as gain media. These materials not only have high material costs, but also require inherently complex integration with Si-based fabrication processing, increasing overall costs and thereby limiting their large-scale adoption. Furthermore, these material-based nanowire lasers rarely emit above 2 um, which is a technologically important wavelength regime for various applications in imaging and quantum sensing. Recently, group-IV nanowires, particularly direct bandgap GeSn nanowires capable of emitting above 2 um, have emerged as promising cost-effective gain media for Si-compatible nanowire lasers, but there has been no successful demonstration of lasing from this seemingly promising nanowire platform. Herein, we report the experimental observation of lasing above 2 um from a single bottom-up grown GeSn nanowire. By harnessing strain engineering and optimized cavity designs simultaneously, the single GeSn nanowire achieves an amplified material gain that can sufficiently overcome minimized optical losses, resulting in a single-mode lasing with an ultra-low threshold of ~5.3 kW cm-2. Our finding paves the way for all-group IV mid-infrared photonic-integrated circuits with compact Si-compatible lasers for on-chip classical and quantum sensing and free-space communication.

physics.optics↗

arXiv2024

Distribution-Adaptive Dynamic Shot Optimization for Variational Quantum Algorithms

Variational quantum algorithms (VQAs) have attracted remarkable interest over the past few years because of their potential computational advantages on near-term quantum devices. They leverage a hybrid approach that integrates classical and quantum computing resources to solve high-dimensional problems that are challenging for classical approaches alone. In the training process of variational circuits, constructing an accurate probability distribution for each epoch is not always necessary, creating opportunities to reduce computational costs through shot reduction. However, existing shot-allocation methods that capitalize on this potential often lack adaptive feedback or are tied to specific classical optimizers, which limits their applicability to common VQAs and broader optimization techniques. Our observations indicate that the information entropy of a quantum circuit's output distribution exhibits an approximately exponential relationship with the number of shots needed to achieve a target Hellinger distance. In this work, we propose a distribution-adaptive dynamic shot (DDS) framework that efficiently adjusts the number of shots per iteration in VQAs using the entropy distribution from the prior training epoch. Our results demonstrate that the DDS framework sustains inference accuracy while achieving a ~50% reduction in average shot count compared to fixed-shot training, and ~60% higher accuracy than recently proposed tiered shot allocation methods. Furthermore, in noisy simulations that reflect the error rates of actual IBM quantum systems, DDS achieves approximately a ~30% reduction in the total number of shots compared to the fixed-shot method with minimal degradation in accuracy, and offers about ~70% higher computational accuracy than tiered shot allocation methods.

quant-ph↗

arXiv2024

Automatic Question-Answer Generation for Long-Tail Knowledge

Pretrained Large Language Models (LLMs) have gained significant attention for addressing open-domain Question Answering (QA). While they exhibit high accuracy in answering questions related to common knowledge, LLMs encounter difficulties in learning about uncommon long-tail knowledge (tail entities). Since manually constructing QA datasets demands substantial human resources, the types of existing QA datasets are limited, leaving us with a scarcity of datasets to study the performance of LLMs on tail entities. In this paper, we propose an automatic approach to generate specialized QA datasets for tail entities and present the associated research challenges. We conduct extensive experiments by employing pretrained LLMs on our newly generated long-tail QA datasets, comparing their performance with and without external resources including Wikipedia and Wikidata knowledge graphs.

cs.CL↗

arXiv2023

Extended-SWIR GeSn LEDs with reduced footprint and power consumption

CMOS-compatible short- and mid-wave infrared emitters are highly coveted for the monolithic integration of silicon-based photonic and electronic integrated circuits to serve a myriad of applications in sensing and communications. In this regard, a group IV germanium-tin (GeSn) material epitaxially grown on silicon (Si) emerges as a promising platform to implement tunable infrared light emitters. Indeed, upon increasing the Sn content, the bandgap of GeSn narrows and becomes direct, making this material system suitable for developing an efficient silicon-compatible emitter. With this perspective, microbridge PIN GeSn LEDs with a small footprint of $1,520$ $μ$m$^2$ are demonstrated and their operation performance is investigated. The spectral analysis of the electroluminescence emission exhibits a peak at $2.31$ $μ$m and it red-shifts slightly as the driving current increases. It is found that the microbridge LED operates at a dissipated power as low as $10.8$ W at room temperature and just $3$ W at $80$ K. This demonstrated low operation power is comparable to that reported for LEDs having a significantly larger footprint reaching $10^6$ $μ$m$^2$. The efficient thermal dissipation of the current design helped to reduce the heat-induced optical losses, thus enhancing light emission. Further performance improvements are envisioned through thermal and optical simulations of the microbridge design. The use of GeSnOI substrate for developing a similar device is expected to improve optical confinement for the realization of electrically driven GeSn lasers.

physics.optics↗

arXiv2022

Keypoint based Sign Language Translation without Glosses

Sign Language Translation (SLT) is a task that has not been studied relatively much compared to the study of Sign Language Recognition (SLR). However, the SLR is a study that recognizes the unique grammar of sign language, which is different from the spoken language and has a problem that non-disabled people cannot easily interpret. So, we're going to solve the problem of translating directly spoken language in sign language video. To this end, we propose a new keypoint normalization method for performing translation based on the skeleton point of the signer and robustly normalizing these points in sign language translation. It contributed to performance improvement by a customized normalization method depending on the body parts. In addition, we propose a stochastic frame selection method that enables frame augmentation and sampling at the same time. Finally, it is translated into the spoken language through an Attention-based translation model. Our method can be applied to various datasets in a way that can be applied to datasets without glosses. In addition, quantitative experimental evaluation proved the excellence of our method.

cs.CV↗

arXiv2022

Are Evolutionary Algorithms Safe Optimizers?

We consider a type of constrained optimization problem, where the violation of a constraint leads to an irrevocable loss, such as breakage of a valuable experimental resource/platform or loss of human life. Such problems are referred to as safe optimization problems (SafeOPs). While SafeOPs have received attention in the machine learning community in recent years, there was little interest in the evolutionary computation (EC) community despite some early attempts between 2009 and 2011. Moreover, there is a lack of acceptable guidelines on how to benchmark different algorithms for SafeOPs, an area where the EC community has significant experience in. Driven by the need for more efficient algorithms and benchmark guidelines for SafeOPs, the objective of this paper is to reignite the interest of this problem class in the EC community. To achieve this we (i) provide a formal definition of SafeOPs and contrast it to other types of optimization problems that the EC community is familiar with, (ii) investigate the impact of key SafeOP parameters on the performance of selected safe optimization algorithms, (iii) benchmark EC against state-of-the-art safe optimization algorithms from the machine learning community, and (iv) provide an open-source Python framework to replicate and extend our work.

cs.NE↗

arXiv2021

1D photonic crystal direct bandgap GeSn-on-insulator laser

GeSn alloys have been regarded as a potential lasing material for a complementary metal-oxide-semiconductor (CMOS)-compatible light source. Despite their remarkable progress, all GeSn lasers reported to date have large device footprints and active areas, which prevent the realization of densely integrated on-chip lasers operating at low power consumption. Here, we present a 1D photonic crystal (PC) nanobeam with a very small device footprint of 7 $μm^2$ and a compact active area of ~1.2 $μm^2$ on a high-quality GeSn-on-insulator (GeSnOI) substrate. We also report that the improved directness in our strain-free nanobeam lasers leads to a lower threshold density and a higher operating temperature compared to the compressive strained counterparts. The threshold density of the strain-free nanobeam laser is ~18.2 kW cm$^{ -2}$ at 4 K, which is significantly lower than that of the unreleased nanobeam laser (~38.4 kW cm$^{ -2}$ at 4 K). Lasing in the strain-free nanobeam device persists up to 90 K, whereas the unreleased nanobeam shows a quenching of the lasing at a temperature of 70 K. Our demonstration offers a new avenue towards developing practical group-IV light sources with high-density integration and low power consumption.

physics.optics↗

arXiv2021

Strain-relaxed GeSn-on-Insulator (GeSnOI) microdisks

GeSn alloys offer a promising route towards a CMOS compatible light source and the realization of electronic-photonic integrated circuits. One tactic to improve the lasing performance of GeSn lasers is to use a high Sn content, which improves the directness. Another popular approach is to use a low to moderate Sn content with either compressive strain relaxation or tensile strain engineering, but these strain engineering techniques generally require optical cavities to be suspended in air, which leads to poor thermal management. In this work, we develop a novel dual insulator GeSn-on-insulator (GeSnOI) material platform that is used to produce strain-relaxed GeSn microdisks stuck on a substrate. By undercutting only one insulating layer (i.e., Al2O3), we fabricate microdisks sitting on SiO2, which attain three key properties for a high-performance GeSn laser: removal of harmful compressive strain, decent thermal management, and excellent optical confinement. We believe that an increase in the Sn content of GeSn layers on our platform can allow achieving improved lasing performance.

physics.optics↗

arXiv2021

Pseudo-magnetic field-induced ultra-slow carrier dynamics in periodically strained graphene

The creation of pseudo-magnetic fields in strained graphene has emerged as a promising route to allow observing intriguing physical phenomena that would be unattainable with laboratory superconducting magnets. Scanning tunneling spectroscopy experiments have successfully measured the pseudo-Landau levels and proved the existence of pseudo-magnetic fields in various strained graphene systems. These giant pseudo-magnetic fields observed in highly deformed graphene can substantially alter the optical properties of graphene beyond a level that can be feasible with an external magnetic field, but the experimental signatures of the influence of such pseudo-magnetic fields have yet to be unveiled. Here, using time-resolved infrared pump-probe spectroscopy, we provide unambiguous evidence for ultra-slow carrier dynamics enabled by pseudo-magnetic fields in periodically strained graphene. Strong pseudo-magnetic fields of ~100 T created by non-uniform strain in graphene nanopillars are found to significantly decelerate the relaxation processes of hot carriers by more than an order of magnitude. Our finding presents unforeseen opportunities for harnessing the new physics of graphene enabled by pseudo-magnetic fields for optoelectronics and condensed matter physics.

cond-mat.mes-hall↗

arXiv2021

Safe Learning and Optimization Techniques: Towards a Survey of the State of the Art

Safe learning and optimization deals with learning and optimization problems that avoid, as much as possible, the evaluation of non-safe input points, which are solutions, policies, or strategies that cause an irrecoverable loss (e.g., breakage of a machine or equipment, or life threat). Although a comprehensive survey of safe reinforcement learning algorithms was published in 2015, a number of new algorithms have been proposed thereafter, and related works in active learning and in optimization were not considered. This paper reviews those algorithms from a number of domains including reinforcement learning, Gaussian process regression and classification, evolutionary algorithms, and active learning. We provide the fundamental concepts on which the reviewed algorithms are based and a characterization of the individual algorithms. We conclude by explaining how the algorithms are connected and suggestions for future research.

cs.LG↗

arXiv2021

Enhanced GeSn Microdisk Lasers Directly Released on Si

GeSn alloys are promising candidates for complementary metal-oxide-semiconductor (CMOS)-compatible, tunable lasers. Relaxation of residual compressive strain in epitaxial GeSn has recently shown promise in improving the lasing performance. However, the suspended device configuration that has thus far been introduced to relax the strain is destined to limit heat dissipation, thus hindering the device performance. Herein, we demonstrate that strain-free GeSn microdisk laser devices fully released on Si outperform the canonical suspended devices. This approach allows to simultaneously relax the limiting compressive strain while offering excellent thermal conduction. Optical simulations confirm that, despite a relatively small refractive index contrast between GeSn and Si, optical confinement in strain-free GeSn optical cavities on Si is superior to that in conventional strain-free GeSn cavities suspended in the air. Moreover, thermal simulations indicate a negligible temperature increase in our device. Conversely, the temperature in the suspended devices increases substantially reaching, for instance, 120 K at a base temperature of 75 K under the employed optical pumping conditions. Such improvements enable increasing the operation temperature by ~40 K and reducing the lasing threshold by 30%. This approach lays the groundwork to implement new designs in the quest for room temperature GeSn lasers on Si.

physics.optics↗

arXiv2018

Data Synthesis based on Generative Adversarial Networks

Privacy is an important concern for our society where sharing data with partners or releasing data to the public is a frequent occurrence. Some of the techniques that are being used to achieve privacy are to remove identifiers, alter quasi-identifiers, and perturb values. Unfortunately, these approaches suffer from two limitations. First, it has been shown that private information can still be leaked if attackers possess some background knowledge or other information sources. Second, they do not take into account the adverse impact these methods will have on the utility of the released data. In this paper, we propose a method that meets both requirements. Our method, called table-GAN, uses generative adversarial networks (GANs) to synthesize fake tables that are statistically similar to the original table yet do not incur information leakage. We show that the machine learning models trained using our synthetic tables exhibit performance that is similar to that of models trained using the original table for unknown testing cases. We call this property model compatibility. We believe that anonymization/perturbation/synthesis methods without model compatibility are of little value. We used four real-world datasets from four different domains for our experiments and conducted in-depth comparisons with state-of-the-art anonymization, perturbation, and generation techniques. Throughout our experiments, only our method consistently shows a balance between privacy level and model compatibility.

cs.DB↗