arXiv subjects

Taeyoon Kim

Publications and source records attributed to Taeyoon Kim.

At least 19 recordsLinked to original sources

arXiv2026

Architectural scaling tradeoffs in modular 3D bosonic quantum processors

We propose a modular three-dimensional bosonic quantum processor built from repeatable coupled-cavity modules linked by configurable interconnect networks. Using hardware-motivated graph-theoretic measures, we compare nearest-neighbor, hub-based, and hybrid architectures in terms of interconnect count, communication distance, resource concentration, and implementation complexity. Rather than identifying a universally optimal topology, our analysis shows how these architectures redistribute the costs of scaling, including wiring and port requirements, nonlocal communication distance, exposure to shared resources, routing bottlenecks, and scheduling overhead. Case studies of a $3\times3$ processor and a larger hierarchical architecture further distinguish finite-size performance from asymptotic scaling. The resulting framework provides a systematic basis for evaluating modular three-dimensional bosonic processors and for identifying the device-level parameters required for quantitative hardware design.

quant-ph↗

arXiv2026

Vivace: Exact Temporal OLAP over Interval Histories via Independent Serverless Execution

Temporal online analytical processing (OLAP) analyzes past states of data whose values change over time. Such histories are naturally stored as interval histories, in which each row records the period during which a value remained valid. Because temporal analyses typically arrive in infrequent, intermittent bursts, serverless execution that launches functions only at query time offers a cost advantage over always-on clusters. Splitting a computation that a single process performs as a whole across independent serverless functions, however, breaks correctness in two ways. A function may not receive the rows that determine the state of its time range, and naively summing partial results yields incorrect answers for duration-weighted and cumulative-threshold queries. Existing SQL engines and serverless analytics do not address both problems together. This paper presents Vivace, a serverless system for exact temporal OLAP over interval histories. Vivace resolves the two problems in separate stages. Before any query arrives, a pre-query layout step partitions the interval history, replicating boundary-crossing intervals so each function computes its range completely from a single file. At query time, a merge step combines partial results under operator-specific rules. Associative aggregates merge intermediate values, and ranking re-orders candidates within each time range. We prove that this partitioned execution matches single-process computation up to canonical form. Evaluated on AWS Lambda with real-world datasets, Vivace reduces latency and monetary cost by up to 82% and 84%, respectively, against an equivalent SQL baseline that queries the history directly, demonstrating robust generality and efficiency.

cs.DB↗

arXiv2026

KubePACS: Kubernetes Cluster Using Performant, Highly Available, and Cost Efficient Spot Instances

Cloud users aim to minimize cost while maximizing performance by selecting the most suitable instance types for their workloads. To reduce expenses, spot instances have been widely adopted due to their steep discounts compared to on-demand pricing. However, their use introduces reliability risks due to potential interruptions, and existing research has primarily focused on mitigating this trade-off from a cost or availability perspective alone. Despite the diversity in hardware capabilities among instance types, current provisioning systems tend to ignore performance variation, selecting nodes solely based on minimum resource requirements. In this paper, we present KubePACS, a Kubernetes-native spot instance provisioning system that constructs node pools optimized for both cost and performance while guaranteeing high availability. KubePACS formulates the node selection process as a multi-objective optimization problem, incorporating real-time data such as spot prices, performance benchmarks, and availability scores, including the multi-node Spot Placement Score (SPS). It solves this problem efficiently using an Integer Linear Programming (ILP) approach guided by the Golden Section Search (GSS) algorithm to find the optimal configuration. By integrating with the Karpenter node autoscaler, KubePACS jointly optimizes instance-type selection and node scaling decisions within a standard provisioning workflow. KubePACS also adopts a novel heuristic to support workload-specific preferences by scaling performance metrics for specialized instances. Through extensive evaluation across synthetic and real-world workloads, KubePACS demonstrates on average 55.09% and up to 81.06% higher performance per dollar over state-of-the-art solutions such as Karpenter, SpotVerse, and SpotKube, which only reference the spot instance prices and limited availability data.

cs.DC↗

arXiv2026

SpotVista: Availability-Aware Recommendation System for Reliable and Cost-Efficient Multi-Node Spot Instances

Cloud vendors offer discounted spot instances to maximize surplus resource utilization, but these instances are subject to the risk of sudden interruption. Traditional pricing datasets have been employed to predict this risk, yet recent policy changes by cloud vendors have diminished their effectiveness. To promote spot instance usage, public cloud vendors provide instant availability datasets to help users mitigate interruption risks. While existing research utilizing this data has proposed methods to reduce interruptions, these studies have primarily focused on single-node instances, overlooking the stability of multi-node environments widely adopted for modern cloud workloads. This paper proposes SpotVista, a system that recommends a resource pool of reliable and cost-efficient multi-node spot instances by leveraging various publicly available datasets. To achieve this, SpotVista collects a large-scale multi-node availability dataset while overcoming significant query limitations. Through a thorough analysis of multi-node spot instance availability behavior, SpotVista establishes a methodology for recommending cost-efficient and reliable multi-node configurations. To evaluate how effectively the proposed methodology reflects multi-node availability and cost efficiency, extensive real-world interruption experiments were conducted. The results demonstrate that SpotVista outperforms the state-of-the-art work, SpotVerse, achieving 81.28% greater availability and 2.84\% more cost savings in a multi-region setup. When compared to a publicly available service, AWS SpotFleet, SpotVista provides 21.6\% higher stability and 26.3% greater cost savings.

cs.DC↗

arXiv2026

Refined 3D index

We introduce a refined version of the 3D index for 3-manifolds, building on the construction of the 3D $\mathcal{N}=2$ gauge theory $T[M]$ by Dimofte-Gaiotto-Gukov and Gang-Yonekura. The refined index is a superconformal index of $T[M]$ equipped with additional gradings that capture enhanced flavor symmetries of the effective theory. Our construction is based on a Dehn surgery presentation of $M$ in terms of an ideally triangulated link complement $N$. We derive an explicit infinite-sum formula for the refined index and provide nontrivial checks in representative examples, supporting its invariance under changes of triangulation, Dehn surgery presentation, and other auxiliary data. As a strictly stronger invariant, the refined index enables finer distinctions among 3-manifolds and among distinct IR phases of the associated gauge theories. We also introduce a computational tool, \textsc{Refined Index Calculator}, for its explicit evaluation.

hep-th↗

arXiv2026

Ding-Dong Ditch: Peeking Into Spot Instance Availability

Spot instances offer significant cost savings of up to 90% over on-demand prices, making them an attractive resource for large-scale computing workloads. However, understanding their availability dynamics is essential for building systems that tolerate interruptions, and observing this availability directly requires keeping instances running, which incurs costs that scale with the number of monitored instance types and their per-instance price. We propose Ding-Dong Ditch (DDD), a cost-efficient method that collects spot instance availability signals by leveraging the cloud provider's provisioning lifecycle. Since the outcome of a spot request is determined before the instance enters the running state, DDD submits requests and cancels them upon provisioning acceptance, collecting binary availability signals at near-zero instance cost. Submitting multiple concurrent requests per measurement point further yields a quantitative estimate of available capacity. We validate DDD through simultaneous collection of probing signals and actual running instance traces across 68 instance types and 15 regions on both AWS and Azure, totaling 336,033 spot requests. Analysis of 2,635 real-world interruption events reveals that co-interruptions within the same instance type and availability zone occur within three minutes in over 92% of cases, motivating a binary availability formulation. Based on this formulation, we derive three complementary features from DDD signals and demonstrate that their combination achieves an F1-macro score of up to 0.90 for current availability modeling and maintains 0.85 at a 60-minute prediction horizon. A trace-driven simulation using TPC-DS workloads further demonstrates the potential of DDD-based prediction to reduce lost computation compared to an unguided baseline.

cs.DC↗

arXiv2026

Why Do AI Agents Systematically Fail at Cloud Root Cause Analysis?

Failures in large-scale cloud systems incur substantial financial losses, making automated Root Cause Analysis (RCA) essential for operational stability. Recent efforts leverage Large Language Model (LLM) agents to automate this task, yet existing systems exhibit low detection accuracy even with capable models, and current evaluation frameworks assess only final answer correctness without revealing why the agent's reasoning failed. This paper presents a process level failure analysis of LLM-based RCA agents. We execute the full OpenRCA benchmark across five LLM models, producing 1,675 agent runs, and classify observed failures into 12 pitfall types across intra-agent reasoning, inter-agent communication, and agent-environment interaction. Our analysis reveals that the most prevalent pitfalls, notably hallucinated data interpretation and incomplete exploration, persist across all models regardless of capability tier, indicating that these failures originate from the shared agent architecture rather than from individual model limitations. Controlled mitigation experiments further show that prompt engineering alone cannot resolve the dominant pitfalls, whereas enriching the inter-agent communication protocol reduces communication-related failures by up to 15 percentage points. The pitfall taxonomy and diagnostic methodology developed in this work provide a foundation for designing more reliable autonomous agents for cloud RCA.

cs.AI↗

arXiv2025

Dynamics of Socio-Institutional Asynchrony in Generative AI: Analyzing the Relative Importance of Intervention Timing vs. Enforcement Efficiency via the Socio-Institutional Asynchrony Model (SIAM)

The super-exponential growth of generative AI has intensified the institutional mismatch between the pace of technological diffusion and the speed of institutional adaptation. This study proposes the Socio-Institutional Asynchrony Model, or SIAM, to quantitatively evaluate the relative effectiveness of two policy levers: intervention timing and enforcement efficiency. Using the timeline of the EU AI Act and an assumed compute doubling time of six months, we conduct a high precision simulation with 10001 time steps. The results show that an earlier intervention timing reduces the cumulative social burden by approximately sixty four percent, whereas improving enforcement efficiency reduces it by only about thirty percent. We further demonstrate analytically that advancing the start of intervention has structurally higher sensitivity, with roughly twice the relative effectiveness, compared to accelerating enforcement speed. These findings suggest that the core value of AI governance lies in proactive timeliness rather than reactive administrative efficiency.

cs.CY↗

arXiv2025

Quantum Enhanced Dark-Matter Search with Entangled Fock States in High-Quality Cavities

We present a quantum-enhanced protocol for detecting wave-like dark matter using an array of $N$ entangled superconducting cavities initialized in an $m$-photon Fock state. By distributing and recollecting the quantum state with an entanglement-distribution operation, the scan rate scales as $N^2(m+1)$ while thermal excitation is the dominant background, significantly outperforming classical single-cavity methods under matched conditions. We evaluate the robustness of our scheme against additional noise sources, including decoherence and beamsplitter infidelity, through theoretical analysis and numerical simulations. In practice, the key requirements, namely high-Q superconducting radio-frequency cavities that support long integration times, high-fidelity microwave beamsplitters, and universal cavity control, are already available on current experimental platforms, making the protocol experimentally feasible.

quant-ph↗

arXiv2025

On-Demand Inverse Design for Narrowband Nanophotonic Structures Based on Generative Model and Tandem Network

Inverse design in nanophotonics remains challenging due to its ill-posed nature and sensitivity to input inaccuracies. We present a novel framework that combines a Conditional Variational Autoencoder (CVAE) with a tandem network, enabling robust and efficient on-demand inverse design of photonic structures. Unlike prior approaches that use CVAEs or tandem networks in isolation, our method integrates spectral adjustment and structural prediction in a unified architecture. Specifically, the CVAE adjusts the idealized target spectra, such as Lorentzian-shaped notches, making them more physically realizable and consistent with the training data distribution. This adjusted spectrum is then passed to the tandem network, which predicts the corresponding structural parameters. The framework effectively handles both narrowband (<50 nm) and highly complex spectra, while addressing the one-to-many mapping challenge inherent in inverse design. The model achieves high accuracy, and the designed spectra closely match full-wave simulation results, validating its practicality for advanced nanophotonic applications.

physics.optics↗

arXiv2025

Ultracoherent superconducting cavity-based multiqudit platform with error-resilient control

Realizing the promise of quantum computing requires simultaneously increasing Hilbert-space size and coherence times. While qubit-based architectures have long been the dominant paradigm, large local Hilbert spaces enable qudits, which offer more compact circuits and natural representations for quantum simulation in chemistry, condensed matter, and high-energy physics. Superconducting radio-frequency (SRF) cavities are attractive building blocks for qudit-based quantum technologies because of their exceptionally low dissipation and large bosonic Hilbert spaces, but turning them into programmable modules requires extra effort because the nonlinear circuitry required for control and measurement often introduces loss and noise that erode the memory advantage. Here we demonstrate a two-mode SRF cavity module weakly coupled to an ancillary transmon circuit and engineered to suppress controller-induced dissipation and dephasing, achieving single-photon lifetimes of 20.6 ms and 15.6 ms and a dephasing time exceeding 40 ms. Using sideband interactions together with error-resilient protocols incorporating measurement-based correction and post-selection, we prepare Fock states up to $N=20$ with fidelities exceeding 95% and generate two-mode entanglement near the coherence limit, approaching 99.9%. By combining ultracoherent storage with high-fidelity control, this work moves cavity-based hardware beyond memory-only operation and establishes a practical route toward high-dimensional encodings and scalable modular quantum information processing.

quant-ph↗

arXiv2024

Demonstrating the Potential of Adaptive LMS Filtering on FPGA-Based Qubit Control Platforms for Improved Qubit Readout in 2D and 3D Quantum Processing Units

Advancements in quantum computing underscore the critical need for sophisticated qubit readout techniques to accurately discern quantum states. This abstract presents our research intended for optimizing readout pulse fidelity for 2D and 3D Quantum Processing Units (QPUs), the latter coupled with Superconducting Radio Frequency (SRF) cavities. Focusing specifically on the application of the Least Mean Squares (LMS) adaptive filtering algorithm, we explore its integration into the FPGA-based control systems to enhance the accuracy and efficiency of qubit state detection by improving Signal-to-Noise Ratio (SNR). Implementing the LMS algorithm on the Zynq UltraScale+ RFSoC Gen 3 devices (RFSoC 4x2 FPGA and ZCU216 FPGA) using the Quantum Instrumentation Control Kit (QICK) open-source platform, we aim to dynamically test and adjust the filtering parameters in real-time to characterize and adapt to the noise profile presented in quantum computing readout signals. Our preliminary results demonstrate the LMS filter's capability to maintain high readout accuracy while efficiently managing FPGA resources. These findings are expected to contribute to developing more reliable and scalable quantum computing architectures, highlighting the pivotal role of adaptive signal processing in quantum technology advancements.

quant-ph↗

arXiv2024

Mechanics and morphology of proliferating cell collectives with self-inhibiting growth

We study the dynamics of proliferating cell collectives whose microscopic constituents' growth is inhibited by macroscopic growth-induced stress. Discrete particle simulations of a growing collective show the emergence of concentric-ring patterns in cell size whose spatio-temporal structure is closely tied to the individual cell's stress response. Motivated by these observations, we derive a multiscale continuum theory whose parameters map directly to the discrete model. Analytical solutions of this theory show the concentric patterns arise from anisotropically accumulated resistance to growth over many cell cycles. This work shows how purely mechanical processes can affect the internal patterning and morphology of cell collectives, and provides a concise theoretical framework for connecting the micro- to macroscopic dynamics of proliferating matter.

physics.bio-ph↗

arXiv2024

Crosstalk-Robust Quantum Control in Multimode Bosonic Systems

High-coherence superconducting cavities offer a hardware-efficient platform for quantum information processing. To achieve universal operations of these bosonic modes, the requisite nonlinearity is realized by coupling them to a transmon ancilla. However, this configuration is susceptible to crosstalk errors in the dispersive regime, where the ancilla frequency is Stark-shifted by the state of each coupled bosonic mode. This leads to a frequency mismatch of the ancilla drive, lowering the gate fidelities. To mitigate such coherent errors, we employ quantum optimal control to engineer ancilla pulses that are robust to the frequency shifts. These optimized pulses are subsequently integrated into a recently developed echoed conditional displacement (ECD) protocol for executing single- and two-mode operations. Through numerical simulations, we examine two representative scenarios: the preparation of single-mode Fock states in the presence of spectator modes and the generation of two-mode entangled Bell-cat states. Our approach markedly suppresses crosstalk errors, outperforming conventional ancilla control methods by orders of magnitude. These results provide guidance for experimentally achieving high-fidelity multimode operations and pave the way for developing high-performance bosonic quantum information processors.

quant-ph↗

arXiv2023

Fast ZZ-Free Entangling Gates for Superconducting Qubits Assisted by a Driven Resonator

Engineering high-fidelity two-qubit gates is an indispensable step toward practical quantum computing. For superconducting quantum platforms, one important setback is the stray interaction between qubits, which causes significant coherent errors. For transmon qubits, protocols for mitigating such errors usually involve fine-tuning the hardware parameters or introducing usually noisy flux-tunable couplers. In this work, we propose a simple scheme to cancel these stray interactions. The coupler used for such cancellation is a driven high-coherence resonator, where the amplitude and frequency of the drive serve as control knobs. Through the resonator-induced-phase (RIP) interaction, the static ZZ coupling can be entirely neutralized. We numerically show that such a scheme can enable short and high-fidelity entangling gates, including cross-resonance CNOT gates within 40 ns and adiabatic CZ gates within 140 ns. Our architecture is not only ZZ free but also contains no extra noisy components, such that it preserves the coherence times of fixed-frequency transmon qubits. With the state-of-the-art coherence times, the error of our cross-resonance CNOT gate can be reduced to below 1e-4.

quant-ph↗

arXiv2023

WHAT, WHEN, and HOW to Ground: Designing User Persona-Aware Conversational Agents for Engaging Dialogue

This paper presents a method for building a personalized open-domain dialogue system to address the WWH (WHAT, WHEN, and HOW) problem for natural response generation in a commercial setting, where personalized dialogue responses are heavily interleaved with casual response turns. The proposed approach involves weighted dataset blending, negative persona information augmentation methods, and the design of personalized conversation datasets to address the challenges of WWH in personalized, open-domain dialogue systems. Our work effectively balances dialogue fluency and tendency to ground, while also introducing a response-type label to improve the controllability and explainability of the grounded responses. The combination of these methods leads to more fluent conversations, as evidenced by subjective human evaluations as well as objective evaluations.

cs.CL↗

arXiv2022

Quantum computing hardware for HEP algorithms and sensing

Quantum information science harnesses the principles of quantum mechanics to realize computational algorithms with complexities vastly intractable by current computer platforms. Typical applications range from quantum chemistry to optimization problems and also include simulations for high energy physics. The recent maturing of quantum hardware has triggered preliminary explorations by several institutions (including Fermilab) of quantum hardware capable of demonstrating quantum advantage in multiple domains, from quantum computing to communications, to sensing. The Superconducting Quantum Materials and Systems (SQMS) Center, led by Fermilab, is dedicated to providing breakthroughs in quantum computing and sensing, mediating quantum engineering and HEP based material science. The main goal of the Center is to deploy quantum systems with superior performance tailored to the algorithms used in high energy physics. In this Snowmass paper, we discuss the two most promising superconducting quantum architectures for HEP algorithms, i.e. three-level systems (qutrits) supported by transmon devices coupled to planar devices and multi-level systems (qudits with arbitrary N energy levels) supported by superconducting 3D cavities. For each architecture, we demonstrate exemplary HEP algorithms and identify the current challenges, ongoing work and future opportunities. Furthermore, we discuss the prospects and complexities of interconnecting the different architectures and individual computational nodes. Finally, we review several different strategies of error protection and correction and discuss their potential to improve the performance of the two architectures. This whitepaper seeks to reach out to the HEP community and drive progress in both HEP research and QIS hardware.

quant-ph↗

arXiv2018

Stochastic Doubly Robust Gradient

When training a machine learning model with observational data, it is often encountered that some values are systemically missing. Learning from the incomplete data in which the missingness depends on some covariates may lead to biased estimation of parameters and even harm the fairness of decision outcome. This paper proposes how to adjust the causal effect of covariates on the missingness when training models using stochastic gradient descent (SGD). Inspired by the design of doubly robust estimator and its theoretical property of double robustness, we introduce stochastic doubly robust gradient (SDRG) consisting of two models: weight-corrected gradients for inverse propensity score weighting and per-covariate control variates for regression adjustment. Also, we identify the connection between double robustness and variance reduction in SGD by demonstrating the SDRG algorithm with a unifying framework for variance reduced SGD. The performance of our approach is empirically tested by showing the convergence in training image classifiers with several examples of missing data.

cs.LG↗