arXiv subjects

Jisu Kim

Publications and source records attributed to Jisu Kim.

At least 19 recordsLinked to original sources

arXiv2026

Minimax Rate for Diffusion-Guided Local Intrinsic Dimension Estimation based on Fokker-Planck Equation

Diffusion-guided methods have recently emerged as effective tools for probing the intrinsic geometry of high-dimensional data. Understanding their accuracy requires quantifying both (i) finite-scale approximation error relative to the true local intrinsic dimension (LID) and (ii) estimation error from finite samples. We address these two questions for the FLIPD field (Kamkari et al., 2024; arXiv:2406.03537), defined at the population level through the logarithmic scale derivative of a Gaussian-smoothed density and admitting a Fokker--Planck representation. For $α$-smooth densities on a regular $d$-dimensional manifold, we establish a uniform $O(σ^2)$ approximation bound relative to the manifold dimension $d$. Under expected integrated squared loss, we prove a minimax lower bound of order $(nσ^d)^{-1}$ for estimating the FLIPD field from $n$ observations when $n^{-1/(2α+d)}\leσ\leσ_0$. A plug-in FLIPD estimator based on the empirical denoising score matching (DSM) solution attains a matching upper bound. Thus the minimax rate is of order $(nσ^d)^{-1}$ and it becomes $n^{-2α/(2α+d)}$ at its smallest scale.

stat.ML↗

arXiv2026

Nonparametric In-Context Learning under Growing Geometric Complexity: Minimax Optimality and Local Geometry-Adaptivity of Transformers

Transformers have become a central architecture for in-context learning (ICL), particularly through their state-of-the-art performance in large language models. This success motivates understanding how transformers exploit intrinsic geometry of real-world data. However, existing nonparametric ICL theory has largely focused on simplified settings, such as Euclidean domains or single manifold models, failing to fully explain its phenomenal empirical success in complex tasks. To address this gap, we study the prediction problem under unknown heterogeneous geometry, modeled by sample size-dependent mixtures of manifolds with heterogeneous dimensions, smoothness, and sampling masses. Under the growing geometric complexity model, we establish a minimax lower bound capturing the aggregate difficulty of the components and construct an oracle tangent local polynomial estimator with a matching upper bound. This estimator is connected to a structure-informed, two-stage softmax transformer with a geometric preconditioner and local polynomial solvers. The transformer achieves negligible approximation error relative to the minimax rate with logarithmic depth and polynomial size. Finally, we derive an in-context generalization bound for near empirical risk minimizers (ERM) over this class. Together, these results identify conditions under which the resulting predictor exploits local geometry and attains minimax rate.

stat.ML↗

arXiv2026

Field demonstration of a self-contained 10 km mobile-to-ground quantum network node via common-aperture co-propagation

In this work, we demonstrate a dual-function, common-aperture free-space optical system that simultaneously supports decoy-state BB84 quantum key distribution (QKD) at 803 nm and bidirectional classical laser communication at 640/660 nm over a 10.31 km terrestrial link from a mobile platform to a stationary ground station. To the best of our knowledge, this is the first field demonstration of a fully integrated common-aperture architecture where a co-propagating classical optical link simultaneously performs full QKD post-processing and ecrypted payload transmission over a 10 km atmospheric channel. The classical link utilizes Manchester-coded on-off keying (MC-OOK) at 10 MHz, achieving a packet error rate of $\sim$ 0.4 $\%$ (estimated bit error rate (BER) of 3.13 $\times$ 10$^{-5}$) with hardware-based self-synchronization and error-packet retransmission. Over the 10.31 km link, the QKD system achieved an average quantum bit error rate (QBER) of 2.512 $\%$ and an average secret key rate of 644 bps using rate-adaptive low-density parity-check (LDPC) codes. Finally, using the distilled keys, real-time AES-256 encrypted image transmission was executed over the co-aligned classical optical channel, proving the feasibility of a completely self-contained, jam resistant mobile quantum optical network.

quant-ph↗

arXiv2026

Quantum key distribution over a 2 km free-space channel with a high secure key rate

Free-space quantum key distribution (QKD) provides crucial advantages, including mobility and deployment exibility, for securing next-generation communication networks. However, atmospheric turbulence and beam wander severely degrade single-photon coupling effciency and link stability over outdoor channels. In this work, we experimentally demonstrate a free-space decoy-state BB84 QKD system operating at 803 nm with a 100 MHz repetition rate over a 2 km outdoor link. By employing an active beam-wander correction based on fast-steering mirrors (FSMs) and position sensitive detectors (PSDs) with a 500 Hz control loop, we enhance the ?ber coupling effciency and achieve an average ?nite-key secure key rate of 110 kbps with an average quantum bit error rate (QBER) of 3.28 %, representing a 2.5 times secure key rate enhancement under an active beam stabilization. This work demonstrates the indispensability of active beam stabilization and provides an operational reference for robust, long-range free-space quantum communication.

quant-ph↗

arXiv2026

Semi-Supervised Kernel Ridge Fréchet Regression

Fréchet regression provides a framework for predicting random object-valued responses by minimizing the conditional expected squared distances. We propose semi-supervised kernel ridge Fréchet regression (SS-KRFR) to exploit unlabeled covariates when labeled samples are scarce. The method learns kernel spectral features from both labeled and unlabeled covariates, then utilizes labeled responses to estimate the conditional Fréchet function by kernel ridge regression. We derive prediction error bounds that distinguish the roles of labeled and unlabeled samples, and establish polynomial convergence rates under suitable conditions. Under additional kernel assumptions, a sharper analysis relaxes the sample size requirements for learning the features. We also develop hybrids that combine features, predictions, or both from models fitted with and without unlabeled covariates, establishing uniform stability over mixing weights in Hadamard spaces at common tuning and a separate rate guarantee for prediction mixing. Simulations and a real-data application show semi-supervised gains and improvements over competing Fréchet regression methods in several settings.

stat.ME↗

arXiv2026

An Extensible Agentic Framework for Expert-Level Automation of Atomistic Simulations

Traditionally, atomistic simulation has been constrained by the computational scaling limits of ab initio methods and the parameterization overhead of empirical force fields. The recent emergence of universal machine learning interatomic potentials has significantly mitigated these bottlenecks, offering near-quantum accuracy and generalizability across diverse chemical spaces at a fraction of the computational cost. However, this shift has relocated the bottleneck to the human dimension: time-consuming mechanical processes, such as input preparation and data analysis, now dominate the research lifecycle. We introduce Paimon, a Platform for Agentic Integration in Materials Optimization and Nanoscale-simulations. Through hundreds of trials on an expert-level liquid electrolyte simulation, we show that Paimon substantially improves the reliability of agentic workflows by suppressing silent errors: plausible yet physically incorrect results. Paimon accommodates a new capability without modifying the agents already present, which we demonstrate by integrating an externally developed scientific agent. As an agent harness for atomistic simulations, Paimon affords researchers a continuous, science-centric workflow throughout the entire simulation lifecycle.

cond-mat.mtrl-sci↗

arXiv2026

A Two-Sample Test on Weighted Persistence Intensity Functions in Topological Data Analysis

Persistence intensity functions provide interpretable and informative first-order summaries of random persistence diagram distributions. We study two-sample testing for equality of persistence intensity functions, allowing the underlying diagram distributions to differ under the null. We construct a weighted-kernel statistic as an unbiased estimator of the squared reproducing-kernel Hilbert space distance between the corresponding weighted intensity embeddings, and calibrate it by studentization. For shrinking bandwidths, we establish uniform asymptotic normality under the null and thereby obtain asymptotic Type I error control. Its power is characterized in terms of the $L^2$ discrepancy between the weighted intensity functions. To accommodate persistence diagrams with possibly unbounded cardinality, we introduce regularity conditions that control the effect of cardinality variation and yield the desired moment bounds for the test statistic. We further show that every probability density on $\{(x,y)\in\mathbb{R}^2:y>x > 0\}$ can be realized as the persistence intensity function of a random diagram. Using these results, we establish that the proposed test attains minimax-optimal separation rates over anisotropic Sobolev balls. Lastly, since the optimal bandwidth is not directly accessible in practice, we adapt a bandwidth aggregation framework.

math.ST↗

arXiv2026

Topological Periodicity Test (TopPT) via Confidence Bound of Time-Delay Embeddings

Time-delay embedding is a fundamental technique in Topological Data Analysis (TDA) for reconstructing phase-space dynamics of time-series data, where persistent homology can reveal loops associated with periodicity. However, rigorous statistical uncertainty quantification for these features remains underdeveloped. First, we analyze the topology of time-delay embeddings, showing that the embedded trajectory is homotopy equivalent to a circle ($S^1$) for periodic signals and contractible for non-periodic ones. We also prove a positive lower bound on the embedding reach, ensuring stable topological features. Second, we develop a subsampling approach to construct confidence bounds for persistence diagrams. Under standard manifold regularity conditions, we derive data-dependent bounds with asymptotic guarantees. Finally, we propose Topological Periodicity Test (TopPT), a hypothesis testing framework for periodicity with asymptotically controlled type I and type II error rates. Experiments on bounded-error synthetic data show that raw TDA detects periodic alternatives while avoiding false rejections on structured non-periodic signals, and that the robust rule is conservative after interpolation-error correction. On PhysioNet Fantasia and BIDMC respiratory waveforms, TDA detects most records but only a subset of local windows, unlike scalar periodogram baselines.

math.ST↗

arXiv2026

Agentic programs: an emerging form of scientific software in computational materials science

Computational materials science has traditionally delegated algorithmic tasks to computers while leaving scientific judgments to humans. We argue that recent LLM-based agent harnesses enable an emerging form of scientific software, agentic programs, that combine deterministic algorithms with bounded LLM-based judgment, task-specific verification, episodic maturation, and complete delegation in production. We illustrate this concept with DeMARS, an agentic program for constructing atomistic models from experimentally measured disordered crystal structures.

cond-mat.mtrl-sci↗

arXiv2026

Quality-Aware Multimodal Fusion Reveals Implicit Identity in Valence-Arousal Features

Conventional face recognition relies on static appearance cues and degrades in unconstrained settings with expression variation, occlusion, and poor lighting. We hypothesize that audiovisual expression dynamics carry identity-discriminative information complementary to static appearance, and that extracting this signal requires multimodal representations robust to the variable input quality of in-the-wild video. To learn such representations, we cast multimodal valence-arousal (VA) estimation as a pretext task and propose Quality-Aware Adaptive Fusion (QAAF), which estimates per-sample, per-modality reliability and adapts each modality's contribution through learned soft gating and a quality-dependent dropout. For the problem of VA estimation, QAAF achieves an average Concordance Correlation Coefficient (CCC) of 0.472 via late fusion ensembling on Aff-wild2, improving over a baseline ensemble under the same setting (0.415) as well as a single-backbone baseline (0.288). Furthermore, the proposed QAAF demonstrates greater resilience to unavailable modalities, with only a 7.5-34.4% relative decrease in CCC when one modality is missing. We then probe whether these VA-trained features encode identity without identity-specific training. On AFEW-VA (67 actors) and YTF (1,595 subjects), VA-trained backbone features rank first among evaluated soft biometric methods, and score-level fusion with ArcFace lowers EER on both datasets (0.022 to 0.021 on AFEW-VA, 0.106 to 0.104 on YTF), correcting 68.2% of ArcFace's false accepts on AFEW-VA. These findings establish multimodal VA estimation as a soft biometric modality complementary to conventional face recognition.

cs.CV↗

arXiv2026

Semi-Supervised Learning with Noisy Proxy Covariates: Generalization Bounds and Distribution Regression

In many modern machine learning pipelines, abundant pretrained representations serve as noisy proxy covariates, while task-specific labels remain scarce. We study semi-supervised regression in this setting, and propose a simple two stage estimator that learns kernel eigenfeatures from all proxy covariates and fits a ridge predictor on labeled data. We derive finite sample bounds showing that fast labeled sample rates are recovered when proxy perturbation is controlled and unlabeled proxy covariates are sufficiently abundant. We also show that distribution regression is a direct special case, with analogous guarantees when the finite bag size is large enough. Experiments show consistent gains over supervised and semi-supervised baselines, especially in low label regimes.

cs.LG↗

arXiv2026

Causal K-Means Clustering

Causal effects are often characterized with population summaries. These might provide an incomplete picture when there are heterogeneous treatment effects across subgroups. Since the subgroup structure is typically unknown, it is more challenging to identify and evaluate subgroup effects than population effects. We propose a new solution to this problem: \emph{Causal k-Means Clustering}, which leverages the k-means clustering algorithm to uncover the unknown subgroup structure. Our problem differs significantly from the conventional clustering setup since the variables to be clustered are unknown counterfactual functions. We present a plug-in estimator which is simple and readily implementable using off-the-shelf algorithms, and study its rate of convergence. We also develop a new bias-corrected estimator based on nonparametric efficiency theory and double machine learning, and show that this estimator achieves fast root-n rates and asymptotic normality in large nonparametric models. Our proposed methods are especially useful for modern outcome-wide studies with multiple treatment levels. Further, our framework is extensible to clustering with generic pseudo-outcomes, such as partially observed outcomes or otherwise unknown functions. Finally, we explore finite sample properties via simulation, and illustrate the proposed methods using a study of mobile-supported self-management for chronic low back pain.

stat.ME↗

arXiv2025

Optimizing Cross-Domain Transfer for Universal Machine Learning Interatomic Potentials

Accurate yet transferable machine-learning interatomic potentials (MLIPs) are essential for accelerating materials and chemical discovery. However, most universal MLIPs overfit to narrow datasets or computational protocols, limiting their reliability across chemical and functional domains. We introduce a transferable multi-domain training strategy that jointly optimizes universal and task-specific parameters through selective regularization, coupled with a domain-bridging set (DBS) that aligns potential-energy surfaces across datasets. Systematic ablation experiments show that small DBS fractions (0.1%) and targeted regularization synergistically enhance out-of-distribution generalization while preserving in-domain fidelity. Trained on fifteen open databases spanning molecules, crystals, and surfaces, our model, SevenNet-Omni, achieves state-of-the-art cross-domain accuracy, including adsorption-energy errors below 0.06 eV on metallic surfaces and 0.1 eV on metal-organic frameworks. Despite containing only 0.5% r$^2$SCAN data, SevenNet-Omni reproduces high-fidelity r$^2$SCAN energetics, demonstrating effective cross-functional transfer from large PBE datasets. This framework offers a scalable route toward universal, transferable MLIPs that bridge quantum-mechanical fidelities and chemical domains.

cond-mat.mtrl-sci↗

arXiv2025

Sound Clouds: Exploring ambient intelligence in public spaces to elicit deep human experience of awe, wonder, and beauty

While the ambient intelligence (AmI) systems we encounter in our daily lives, including security monitoring and energy-saving systems, typically serve pragmatic purposes, we wonder how we can design and implement ambient artificial intelligence experiences in public spaces that elicit deep human feelings of awe, wonder, and beauty. As a manifestation, we introduce Sound Clouds, an immersive art installation that generates live music based on participants' interaction with several human-height spheres. Our installation serves as a provocation into future ambient intelligence that provokes, not limits, the future possibilities of AmI.

cs.HC↗

arXiv2025

Memorization or Reasoning? Exploring the Idiom Understanding of LLMs

Idioms have long posed a challenge due to their unique linguistic properties, which set them apart from other common expressions. While recent studies have leveraged large language models (LLMs) to handle idioms across various tasks, e.g., idiom-containing sentence generation and idiomatic machine translation, little is known about the underlying mechanisms of idiom processing in LLMs, particularly in multilingual settings. To this end, we introduce MIDAS, a new large-scale dataset of idioms in six languages, each paired with its corresponding meaning. Leveraging this resource, we conduct a comprehensive evaluation of LLMs' idiom processing ability, identifying key factors that influence their performance. Our findings suggest that LLMs rely not only on memorization, but also adopt a hybrid approach that integrates contextual cues and reasoning, especially when processing compositional idioms. This implies that idiom understanding in LLMs emerges from an interplay between internal knowledge retrieval and reasoning-based inference.

cs.CL↗

arXiv2025

Radio-Frequency Detection of Fabry-Pérot Interference and Quantum Capacitance in Long-Channel Three-Dimensional Dirac Semimetal Cd3As2 Nanowires

We demonstrate phase-coherent transport in suspended long-channel Cd3As2 nanowire devices using both direct current (DC) transport and radio-frequency (RF) reflectometry measurements. By integrating Cd3As2 nanowires with on-chip superconducting LC resonators, we achieve sensitive detection of both resistance and quantum capacitance variations. In a long-channel device (L ~ 1.8 μm), clear Fabry-Pérot (FP) interference patterns are observed in both DC and RF measurements, provide strong evidence for ballistic electron transport. RF reflectometry reveals gate-dependent modulations of the resonance frequency, arising from quantum capacitance oscillations induced by changes in the density of states and FP interference. These oscillations exhibit a quasi-periodic structure that closely correlates with the FP patterns in DC transport measurements. In another device of a Cd3As2 nanowire Josephson junction (L ~ 730 nm, superconducting Al contacts), FP interference patterns are too weak to be resolved in DC conductance but are detectable using RF reflectometry. These results demonstrate the high quality of our Cd3As2 nanowires and the versatility of RF reflectometry, establishing their potential for applications in topological quantum devices, such as Andreev qubits or gatemon architectures.

cond-mat.mes-hall↗

arXiv2025

Chunk Knowledge Generation Model for Enhanced Information Retrieval: A Multi-task Learning Approach

Traditional query expansion techniques for addressing vocabulary mismatch problems in information retrieval are context-sensitive and may lead to performance degradation. As an alternative, document expansion research has gained attention, but existing methods such as Doc2Query have limitations including excessive preprocessing costs, increased index size, and reliability issues with generated content. To mitigate these problems and seek more structured and efficient alternatives, this study proposes a method that divides documents into chunk units and generates textual data for each chunk to simultaneously improve retrieval efficiency and accuracy. The proposed "Chunk Knowledge Generation Model" adopts a T5-based multi-task learning structure that simultaneously generates titles and candidate questions from each document chunk while extracting keywords from user queries. This approach maximizes computational efficiency by generating and extracting three types of semantic information in parallel through a single encoding and two decoding processes. The generated data is utilized as additional information in the retrieval system. GPT-based evaluation on 305 query-document pairs showed that retrieval using the proposed model achieved 95.41% accuracy at Top@10, demonstrating superior performance compared to document chunk-level retrieval. This study contributes by proposing an approach that simultaneously generates titles and candidate questions from document chunks for application in retrieval pipelines, and provides empirical evidence applicable to large-scale information retrieval systems by demonstrating improved retrieval accuracy through qualitative evaluation.

cs.IR↗

arXiv2025

SLM-Based Agentic AI with P-C-G: Optimized for Korean Tool Use

We propose a small-scale language model (SLM) based agent architecture, Planner-Caller-Generator (P-C-G), optimized for Korean tool use. P-C-G separates planning, calling, and generation by role: the Planner produces an initial batch plan with limited on-demand replanning; the Caller returns a normalized call object after joint schema-value validation; and the Generator integrates tool outputs to produce the final answer. We apply a Korean-first value policy to reduce execution failures caused by frequent Korean-to-English code switching in Korean settings. Evaluation assumes Korean queries and Korean tool/parameter specifications; it covers single-chain, multi-chain, missing-parameters, and missing-functions scenarios, and is conducted via an LLM-as-a-Judge protocol averaged over five runs under a unified I/O interface. Results show that P-C-G delivers competitive tool-use accuracy and end-to-end quality while reducing tokens and maintaining acceptable latency, indicating that role-specialized SLMs are a cost-effective alternative for Korean tool-use agents.

cs.CL↗