arXiv subjects

Donghun Lee

Publications and source records attributed to Donghun Lee.

35 records · Page 2Linked to original sources

arXiv2025

CLIP-KOA: Enhancing Knee Osteoarthritis Diagnosis with Multi-Modal Learning and Symmetry-Aware Loss Functions

Knee osteoarthritis (KOA) is a universal chronic musculoskeletal disorders worldwide, making early diagnosis crucial. Currently, the Kellgren and Lawrence (KL) grading system is widely used to assess KOA severity. However, its high inter-observer variability and subjectivity hinder diagnostic consistency. To address these limitations, automated diagnostic techniques using deep learning have been actively explored in recent years. In this study, we propose a CLIP-based framework (CLIP-KOA) to enhance the consistency and reliability of KOA grade prediction. To achieve this, we introduce a learning approach that integrates image and text information and incorporate Symmetry Loss and Consistency Loss to ensure prediction consistency between the original and flipped images. CLIP-KOA achieves state-of-the-art accuracy of 71.86\% on KOA severity prediction task, and ablation studies show that CLIP-KOA has 2.36\% improvement in accuracy over the standard CLIP model due to our contribution. This study shows a novel direction for data-driven medical prediction not only to improve reliability of fine-grained diagnosis and but also to explore multimodal methods for medical image analysis. Our code is available at https://github.com/anonymized-link.

cs.CV↗

arXiv2025

Quantum sensing with arbitrary frequency resolution via correlation measurements

Achieving high-frequency spectral resolution with quantum sensors, while crucial in fields ranging from physical to biological sciences, is challenging due to their finite coherence time. Here, we introduce a novel protocol that achieves this goal by measuring phase correlations of AC magnetic fields using ensembles of NV centers. Our method extends the sensing dynamic range to frequencies higher than the system's Rabi frequency while achieving arbitrary frequency resolution, limited only by the target field coherence time. Moreover, our approach operates more robustly with respect to the magnetic field's amplitude. Thanks to this robustness, our protocol allows the application of more $π$-pulses in pulse sequences such as CPMG, enabling the decoupling of a broader range of frequency noise. The higher harmonics generated in this process continue to act as a part of the signal, ultimately improving the frequency resolution. This method paves the way for achieving arbitrary frequency resolution with improved performances, making it highly versatile for quantum sensing applications across diverse scientific fields.

quant-ph↗

arXiv2025

Kanana: Compute-efficient Bilingual Language Models

We introduce Kanana, a series of bilingual language models that demonstrate exceeding performance in Korean and competitive performance in English. The computational cost of Kanana is significantly lower than that of state-of-the-art models of similar size. The report details the techniques employed during pre-training to achieve compute-efficient yet competitive models, including high quality data filtering, staged pre-training, depth up-scaling, and pruning and distillation. Furthermore, the report outlines the methodologies utilized during the post-training of the Kanana models, encompassing supervised fine-tuning and preference optimization, aimed at enhancing their capability for seamless interaction with users. Lastly, the report elaborates on plausible approaches used for language model adaptation to specific scenarios, such as embedding, retrieval augmented generation, and function calling. The Kanana model series spans from 2.1B to 32.5B parameters with 2.1B models (base, instruct, embedding) publicly released to promote research on Korean language models.

cs.CL↗

arXiv2024

ABC3: Active Bayesian Causal Inference with Cohn Criteria in Randomized Experiments

In causal inference, randomized experiment is a de facto method to overcome various theoretical issues in observational study. However, the experimental design requires expensive costs, so an efficient experimental design is necessary. We propose ABC3, a Bayesian active learning policy for causal inference. We show a policy minimizing an estimation error on conditional average treatment effect is equivalent to minimizing an integrated posterior variance, similar to Cohn criteria \citep{cohn1994active}. We theoretically prove ABC3 also minimizes an imbalance between the treatment and control groups and the type 1 error probability. Imbalance-minimizing characteristic is especially notable as several works have emphasized the importance of achieving balance. Through extensive experiments on real-world data sets, ABC3 achieves the highest efficiency, while empirically showing the theoretical results hold.

cs.LG↗

arXiv2024

Magnetic steganography based on wide field diamond quantum microscopy

We experimentally demonstrate magnetic steganography using wide field quantum microscopy based on diamond nitrogen vacancy centers. The method offers magnetic imaging capable of revealing concealed information otherwise invisible with conventional optical measurements. For a proof of principle demonstration of the magnetic steganography, micrometer structures designed as pixel arts, barcodes, and QR codes are fabricated using mixtures of magnetic and nonmagnetic materials, nickel and gold. We compare three different imaging modes based on the changes in frequency, linewidth, and contrast of the NV electron spin resonance, and find that the last mode offers the best quality of reconstructing hidden magnetic images. By simultaneous driving of the NV qutrit states with two independent microwave fields, we expediate the imaging time by a factor of three. This work shows potential applications of quantum magnetic imaging in the field of image steganography.

quant-ph↗

arXiv2024

Pre-trained Language Models Return Distinguishable Probability Distributions to Unfaithfully Hallucinated Texts

In this work, we show the pre-trained language models return distinguishable generation probability and uncertainty distribution to unfaithfully hallucinated texts, regardless of their size and structure. By examining 24 models on 6 data sets, we find out that 88-98% of cases return statistically significantly distinguishable generation probability and uncertainty distributions. Using this general phenomenon, we showcase a hallucination-reducing training algorithm. Our algorithm outperforms other baselines by achieving higher faithfulness metrics while maintaining sound general text quality measures.

cs.CL↗

arXiv2024

Block Orthogonal Sparse Superposition Codes for $ \sf{L}^3 $ Communications: Low Error Rate, Low Latency, and Low Power Consumption

Block orthogonal sparse superposition (BOSS) code is a class of joint coded modulation methods, which can closely achieve the finite-blocklength capacity with a low-complexity decoder at a few coding rates under Gaussian channels. However, for fading channels, the code performance degrades considerably because coded symbols experience different channel fading effects. In this paper, we put forth novel joint demodulation and decoding methods for BOSS codes under fading channels. For a fast fading channel, we present a minimum mean square error approximate maximum a posteriori (MMSE-A-MAP) algorithm for the joint demodulation and decoding when channel state information is available at the receiver (CSIR). We also propose a joint demodulation and decoding method without using CSIR for a block fading channel scenario. We refer to this as the non-coherent sphere decoding (NSD) algorithm. Simulation results demonstrate that BOSS codes with MMSE-A-MAP decoding outperform CRC-aided polar codes, while NSD decoding achieves comparable performance to quasi-maximum likelihood decoding with significantly reduced complexity. Both decoding algorithms are suitable for parallelization, satisfying low-latency constraints. Additionally, real-time simulations on a software-defined radio testbed validate the feasibility of using BOSS codes for low-power transmission.

cs.IT↗

arXiv2023

P5: Plug-and-Play Persona Prompting for Personalized Response Selection

The use of persona-grounded retrieval-based chatbots is crucial for personalized conversations, but there are several challenges that need to be addressed. 1) In general, collecting persona-grounded corpus is very expensive. 2) The chatbot system does not always respond in consideration of persona at real applications. To address these challenges, we propose a plug-and-play persona prompting method. Our system can function as a standard open-domain chatbot if persona information is not available. We demonstrate that this approach performs well in the zero-shot setting, which reduces the dependence on persona-ground training data. This makes it easier to expand the system to other languages without the need to build a persona-grounded corpus. Additionally, our model can be fine-tuned for even better performance. In our experiments, the zero-shot model improved the standard model by 7.71 and 1.04 points in the original persona and revised persona, respectively. The fine-tuned model improved the previous state-of-the-art system by 1.95 and 3.39 points in the original persona and revised persona, respectively. To the best of our knowledge, this is the first attempt to solve the problem of personalized response selection using prompt sequences. Our code is available on github~\footnote{https://github.com/rungjoo/plug-and-play-prompt-persona}.

cs.CL↗

arXiv2023

Exploring the magnetic properties of individual barcode nanowires using wide-field diamond microscopy

Barcode magnetic nanowires typically comprise a multilayer magnetic structure in a single body with more than one segment type. Interestingly, owing to selective functionalization and novel interactions between the layers, barcode magnetic nanowires have attracted significant attention, particularly in the field of bioengineering. However, an analysis of their magnetic properties at the individual nanowire level remains challenging. With this background, herein, we investigated the characterization of magnetic nanowires at room temperature under ambient conditions based on magnetic images obtained via wide-field quantum microscopy with nitrogen-vacancy centers in diamond. Consequently, we could extract critical magnetic properties, such as the saturation magnetization and coercivity, of single nanowires by comparing the experimental results with those of micromagnetic simulations. This study opens up the possibility for a versatile characterization method suited to individual magnetic nanowires.

quant-ph↗

arXiv2020

Probing sterile neutrino in $B$ ($D$) meson decays at Belle II (BESIII)

We present, how a systematic study of $B \to D\ell N$ ($D \to K \ell N$) decays with $\ell=μ,τ$, at Belle II (BESIII) can provide unambiguous signature of a heavy neutrino $N$ and/or constrain its mixing with active neutrinos $ν_\ell$, which is parameterized by $|U_{\ell N}|^2$. Our constraint on $|U_{μN}|^2$ that can be achieved from the full Belle II data is comparable with what can be obtained from the much larger data set of the upgraded LHCb. Additionally, our method offers better constraint on $|U_{μN}|^2$ for mass of sterile neutrino $m_N < 2$ GeV. We can also probe the Dirac and Majorana nature of $N$ by observing the sequential decay of $N$, including suppression from observation of a displaced vertex as well as helicity flip, for Majorana $N$.

hep-ph↗

arXiv2020

Probing new physics scenarios of muon $g-2$ via $J/ψ$ decay at BESIII

The disagreement between the standard model prediction and the experimental measurement of muon anomalous magnetic moment can be alleviated by invoking an additional particle which is either a vector boson ($X_1$) or a scalar ($X_0$). This new particle, with the mass $m_X \lesssim 2 m_μ$, can be searched for in the decay $J/ψ\to μ^- μ^+ X$, where $X$ is missing. Our numerical study shows that the search is quite feasible at the BESIII experiment in the parameter space allowed by muon $g-2$ measurements.

hep-ph↗

arXiv2019

Truly Batch Apprenticeship Learning with Deep Successor Features

We introduce a novel apprenticeship learning algorithm to learn an expert's underlying reward structure in off-policy model-free \emph{batch} settings. Unlike existing methods that require a dynamics model or additional data acquisition for on-policy evaluation, our algorithm requires only the batch data of observed expert behavior. Such settings are common in real-world tasks---health care, finance or industrial processes ---where accurate simulators do not exist or data acquisition is costly. To address challenges in batch settings, we introduce Deep Successor Feature Networks(DSFN) that estimate feature expectations in an off-policy setting and a transition-regularized imitation network that produces a near-expert initial policy and an efficient feature representation. Our algorithm achieves superior results in batch settings on both control benchmarks and a vital clinical task of sepsis management in the Intensive Care Unit.

cs.LG↗

arXiv2018

Evaluating Reinforcement Learning Algorithms in Observational Health Settings

Much attention has been devoted recently to the development of machine learning algorithms with the goal of improving treatment policies in healthcare. Reinforcement learning (RL) is a sub-field within machine learning that is concerned with learning how to make sequences of decisions so as to optimize long-term effects. Already, RL algorithms have been proposed to identify decision-making strategies for mechanical ventilation, sepsis management and treatment of schizophrenia. However, before implementing treatment policies learned by black-box algorithms in high-stakes clinical decision problems, special care must be taken in the evaluation of these policies. In this document, our goal is to expose some of the subtleties associated with evaluating RL algorithms in healthcare. We aim to provide a conceptual starting point for clinical and computational researchers to ask the right questions when designing and evaluating algorithms for new ways of treating patients. In the following, we describe how choices about how to summarize a history, variance of statistical estimators, and confounders in more ad-hoc measures can result in unreliable, even misleading estimates of the quality of a treatment policy. We also provide suggestions for mitigating these effects---for while there is much promise for mining observational health data to uncover better treatment policies, evaluation must be performed thoughtfully.

cs.LG↗

arXiv2018

Deep Learning on Key Performance Indicators for Predictive Maintenance in SAP HANA

With a new era of cloud and big data, Database Management Systems (DBMSs) have become more crucial in numerous enterprise business applications in all the industries. Accordingly, the importance of their proactive and preventive maintenance has also increased. However, detecting problems by predefined rules or stochastic modeling has limitations, particularly when analyzing the data on high-dimensional Key Performance Indicators (KPIs) from a DBMS. In recent years, Deep Learning (DL) has opened new opportunities for this complex analysis. In this paper, we present two complementary DL approaches to detect anomalies in SAP HANA. A temporal learning approach is used to detect abnormal patterns based on unlabeled historical data, whereas a spatial learning approach is used to classify known anomalies based on labeled data. We implement a system in SAP HANA integrated with Google TensorFlow. The experimental results with real-world data confirm the effectiveness of the system and models.

cs.LG↗

arXiv2016

Topical Review: Spins and mechanics in diamond

There has been rapidly growing interest in hybrid quantum devices involving a solid-state spin and a macroscopic mechanical oscillator. Such hybrid devices create exciting opportunities to mediate interactions between disparate qubits and to explore the quantum regime of macroscopic mechanical objects. In particular, a system consisting of the nitrogen-vacancy defect center in diamond coupled to a high quality factor mechanical oscillator is an appealing candidate for such a hybrid quantum device, as it utilizes the highly coherent and versatile spin properties of the defect center. In this paper, we will review recent experimental progress on diamond-based hybrid quantum devices in which the spin and orbital dynamics of single defects are driven by the motion of a mechanical oscillator. In addition, we discuss prospective applications for this device, including long range, phonon-mediated spin-spin interactions, and phonon cooling in the quantum regime. We conclude the review by evaluating the experimental limitations of current devices and identifying alternative device architectures that may reach the strong coupling regime.

quant-ph↗

arXiv2016

Strain coupling of a mechanical resonator to a single quantum emitter in diamond

The recent maturation of hybrid quantum devices has led to significant enhancements in the functionality of a wide variety of quantum systems. In particular, harnessing mechanical resonators for manipulation and control has expanded the use of two-level systems in quantum information science and quantum sensing. In this letter, we report on a monolithic hybrid quantum device in which strain fields associated with resonant vibrations of a diamond cantilever dynamically control the optical transitions of a single nitrogen-vacancy (NV) defect center in diamond. We quantitatively characterize the strain coupling to the orbital states of the NV center, and with mechanical driving, we observe NV-strain couplings exceeding 10 GHz. Furthermore, we use this strain-mediated coupling to match the frequency and polarization dependence of the zero-phonon lines of two spatially separated and initially distinguishable NV centers. The experiments demonstrated here mark an important step toward engineering a quantum device capable of realizing and probing the dynamics of non-classical states of mechanical resonators, spin-systems, and photons.

quant-ph↗

arXiv2012

A Semantic-Based Approach for Detecting and Decomposing God Classes

Cohesion is a core design quality that has a great impact on posterior development and maintenance. By the nature of software, the cohesion of a system is diminished as the system evolves. God classes are code defects resulting from software evolution, having heterogeneous responsibilities highly coupled with other classes and often large in size, which makes it difficult to maintain the system. The existing work on identifying and decomposing God classes heavily relies on internal class information to identify God classes and responsibilities. However, in object-oriented systems, responsibilities should be analyzed with respect to not only internal class information, but also method interactions. In this paper, we present a novel approach for detecting God classes and decomposing their responsibilities based on the semantics of methods and method interactions. We evaluate the approach using JMeter v2.5.1 and the results are promising.

cs.SE↗