Searcharxiv⌕ Search

arXiv subjects

Taehyun Kim

Publications and source records attributed to Taehyun Kim.

At least 19 recordsLinked to original sources
arXiv2026

Real-Time Dynamics-Based Torque-Sampling MPPI for Compliant and Force Aware Manipulation

This study proposes a novel Model Predictive Path Integral (MPPI)-based task-space control framework. The proposed framework explicitly solves rigid-body dynamics within a real-time MPC formulation and enforces safety constraints, enabling accurate motion and force control that yields compliant behaviors for safe and effective physical interaction of robotic manipulators in unstructured environments. By leveraging MPPI, the proposed framework efficiently handles nonlinear dynamics that are difficult to solve with conventional MPC approaches in real-time. Furthermore, we develop a torque-sampling-based control architecture that enables efficient exploitation of GPU-based parallelization, resulting in effective compliant and force-aware behaviors. As a result, the proposed framework achieves a solver update rate of over 166 Hz with a 0.18 s prediction horizon, and its performance is validated through real-world experiments on a 7-DoF manipulator.

cs.RO↗
arXiv2026

Evolution of bar-induced dark gaps in galaxy discs: evidence of strong bar-driven effects already at $z > 2$

The properties of stellar bars play a crucial role in determining the bar-driven secular evolution in disc galaxies. However, a systematic observational study of the evolution of several bar properties (such as strength and length) across cosmic time is largely missing. In this paper, using a sample of $625$ barred galaxies, taken from SDSS, HST COSMOS, and JWST CEERS surveys, we systematically investigate the evolution of bar properties over redshifts ($0.02 \lesssim z < 3$) by making a novel usage of dark gap (preferential light deficit along the bar minor axis) properties as a proxy for bar properties. We show that the dark gap strength ($\Delta \mu_{\rm max}$) exhibits a weak evolution, increasing from higher redshifts ($z \sim 2.5$) and slightly declining towards lower redshifts ($z < 0.05$). Conversely, the extent of dark gaps ($R_{\rm DG}, R_{\rm dark}$; normalised by bar length) decreases moderately from $z \geq 1.4$ and remains constant thereafter. Our results suggest that bar formation and the initial rapid growth phase occur before $z \sim 3$, followed by mild growth towards lower redshifts. We also find $R_{\rm dark}$ to be a better proxy (as compared to $R_{\rm DG}$) for estimating bar length, supporting earlier theoretical studies. Furthermore, the $\Delta \mu_{\rm max}$ shows a weak but statistically significant correlation with bar-to-total light ratio (Bar/T) and bar ellipticity ($\epsilon_{\rm bar}$). Studies of the redshift evolution of bar properties over such an extensive redshift range as done here are instrumental in constraining the bar-driven evolution at early cosmic times.

astro-ph.GA↗
arXiv2026

Motif 3: Technical Report

We introduce Motif 3, a decoder-only Mixture-of-Experts language model with 314 billion total parameters and 13.2 billion activated per token. Each sparse MoE layer contains 384 routed experts, with eight selected per token. This fine-grained sparsity provides substantial expert capacity while limiting computation. Motif 3 is built around Grouped Differential Latent Attention (GDLA), which integrates grouped differential attention with the compressed key-value representation of Multi-head Latent Attention. The architecture further incorporates modified manifold-constrained hyper-connections, Expert Specific PolyNorm activations, and multi-token prediction to improve optimization stability, expert specialization, and inference efficiency. We pretrain Motif 3 on approximately 12.5 trillion tokens spanning web documents, STEM, code, mathematics, multilingual content, and domain-specialized corpora. Expert-balancing and numerical-stabilization techniques support stable training at scale, while selective MXFP8 computation and communication, memory-efficient fused kernels, and window-aware context parallelism enable training with context lengths up to 256K tokens. Our post-training pipeline combines general supervised fine-tuning, six specialist teachers trained with reinforcement learning, a software-engineering teacher trained with supervised fine-tuning, and Multi-teacher On-Policy Distillation. The resulting unified model consolidates complementary capabilities in reasoning, coding, tool use, professional work, long-context understanding, calibrated abstention, and instruction following. Across a broad evaluation suite, Motif 3 demonstrates competitive performance against leading open weight models, including strong results on long-horizon agentic tasks, mathematical reasoning, scientific knowledge, and hallucination-sensitive evaluation.

cs.AI↗
arXiv2026

The onset of stellar bars at Cosmic Noon. Bar-driven quenching and AGN co-evolution in a mature disc galaxy

Observations with the JWST revealed an unexpected abundance of barred galaxies at Cosmic Noon. However, the physical properties of these early bars are almost unconstrained, as it is their impact in the structural evolution of high-z disc galaxies. In this work, we derived the stellar populations of EGS-24154, a barred spiral galaxy at $z=1.17$. First, we investigated the role of the stellar bar in the early assembly history and structural evolution of the galaxy. Second, we studied the properties of the interstellar medium to shed light on the interplay with the central supermassive black hole. We analysed medium-resolution NIRSpec/IFS data of EGS-24154 through full-spectral fitting and derived light and mass-weighted ages and metallicities. We then reconstructed the spatially-resolved SFH, derived the ionizing mechanisms of the interstellar medium analysing several emission lines, characterized the dynamics of EGS-24154, and constrained the properties of a biconical outflow launched by the AGN. EGS-24154 is a baryon-dominated, gas-rich disc galaxy, which grew more than 90% of its stellar mass when the Universe was ~2 Gyr old. We found that the stellar population of the bar started to form at $z\sim5$, compatibly to the time when the stellar disc started to assemble. We observed a star formation desert in the bar region, which is responsible for quenching star formation over several Gyr. We then interpreted that the feedback from the AGN prevented the growth of central mass concentration, allowing the stellar bar to grow in size and strength. In this first study of spatially-resolved stellar populations of a barred disc galaxy at $z>1$, we demonstrated how stellar bars are key drivers of the early structural and dynamical evolution of disc galaxies. In particular, our results call for a revision of most models of disc and bar formation in early baryon-dominated, gas-rich disc galaxies.

astro-ph.GA↗
arXiv2026

Poisson Empirical Bayes via Gamma-Smoothed Nonparametric Maximum Likelihood

Empirical Bayes methods are widely used for large-scale estimation and inference in the Poisson means problem. Existing results establish theoretical properties of the nonparametric maximum likelihood estimator (NPMLE) for optimal posterior mean estimation, but comparatively less is known about uncertainty quantification (i.e., construction of confidence sets). Two main challenges in constructing confidence sets for the latent parameters based on the NPMLE are its discreteness and its slow rate of prior estimation. We resolve these limitations by introducing a smooth NPMLE that models the prior as a Gamma mixture, which is a flexible class capable of approximating a wide range of continuous priors on $(0,\infty)$. This procedure preserves the convex optimization structure of the classical NPMLE. The smooth NPMLE achieves the optimal nearly parametric rate for posterior mean estimation. Moreover, it achieves a polynomial convergence rate for prior and posterior density estimation under a compact support assumption on the mixing distribution. Based on the smooth NPMLE, we construct plug-in empirical Bayes confidence sets that mimic the oracle optimal (in terms of expected length) marginal coverage sets. We show theoretically and empirically that these sets achieve asymptotically exact marginal coverage and are substantially shorter than existing methods.

math.ST↗
arXiv2026

The Changing-look Phenomenon Accompanied by an Accretion Mode Transition in NGC 3786

To reveal the physical origin of the changing-look (CL) phenomenon in NGC 3786, which transitioned from type 1.8/1.9 to type 1, we present an analysis of long-term spectral monitoring in the optical and near-infrared obtained with Gemini/GMOS-N and Gemini/GNIRS, respectively. Since the onset of the CL phenomenon, NGC 3786 has remained $\sim 1-1.5$ mag brighter in the mid-infrared than in the pre-CL stage, whereas the optical continuum has changed only moderately ($\sim 0.2-0.3$ mag). Spectroscopic analysis further reveals that while the fluxes of the broad Pa$\beta$ and Pa$\alpha$ lines were enhanced over a two-year follow-up period, the flux of the broad H$\alpha$ line remained unchanged. We propose that observed temporal variations in the continuum and line flux ratios disfavor a tidal disruption event origin. Instead, the observations can be primarily explained by a gradual change in line-of-sight extinction driven by variations in the torus covering factor, which is determined by the Eddington ratio and the accretion mode. An additional mechanism, arising from the physical conditions within the broad-line region, may partially account for the temporal evolution of the flux ratios. Our study highlights the importance of investigating the CL phenomenon in intermediate-type active galactic nuclei associated with outbursts detected only in the mid-infrared to explore the detailed structural evolution of nuclear activity.

astro-ph.GA↗
arXiv2026

Motif-Video 2B: Technical Report

Training strong video generation models usually requires massive datasets, large parameter counts, and substantial compute. In this work, we ask whether strong text-to-video quality is possible at a much smaller budget: fewer than 10M clips and less than 100,000 H200 GPU hours. Our core claim is that part of the answer lies in how model capacity is organized, not only in how much of it is used. In video generation, prompt alignment, temporal consistency, and fine-detail recovery can interfere with one another when they are handled through the same pathway. Motif-Video 2B addresses this by separating these roles architecturally, rather than relying on scale alone. The model combines two key ideas. First, Shared Cross-Attention strengthens text control when video token sequences become long. Second, a three-part backbone separates early fusion, joint representation learning, and detail refinement. To make this design effective under a limited compute budget, we pair it with an efficient training recipe based on dynamic token routing and early-phase feature alignment to a frozen pretrained video encoder. Our analysis shows that later blocks develop clearer cross-frame attention structure than standard single-stream baselines. On VBench, Motif-Video~2B reaches 83.76\%, surpassing Wan2.1 14B while using 7$\times$ fewer parameters and substantially less training data. These results suggest that careful architectural specialization, combined with an efficiency-oriented training recipe, can narrow or exceed the quality gap typically associated with much larger video models.

cs.CV↗
arXiv2026

The Identification of Asymmetric Barred Galaxies in Illustris TNG-50

Most barred galaxies exhibit symmetric structures. However, recent studies have shown that a subset of barred galaxies exhibit lopsided morphologies. To quantify their occurrence and investigate their physical origins, we analyze barred galaxies in the IllustrisTNG TNG50 simulation. We select 519 clearly barred galaxies in their stellar mass maps out of 770 barred galaxies from the TNG50-1 catalog. We classify the bar morphologies into four subgroups - `Lopsided', `Perturbed', `Symmetric', and `Indeterminate' - and perform a comparative analysis of their physical properties. We find that galaxies hosting asymmetric bars (`Lopsided' and `Perturbed') tend to have higher gas densities around the bar region, enhanced star formation activity, and more recent bar-formation epochs than galaxies with symmetric bars. However, the factor that most consistently distinguishes the four subgroups is the stellar mass distribution of the host galaxy, and there appears to be no physical correlation with bar size. These findings suggest that asymmetric bars form preferentially in less massive galaxies and may evolve into symmetric bars over time through secular processes. However, this conclusion should be considered with caution, as the fraction of asymmetric bars in the TNG50 simulation is systematically higher than that observed in the local universe.

astro-ph.GA↗
arXiv2026

Empirical Bayes Estimation and Inference via Smooth Nonparametric Maximum Likelihood

The empirical Bayes $g$-modeling approach based on the nonparametric maximum likelihood estimator (NPMLE) has been central to large-scale estimation and inference in the normal means problem. However, theoretical guarantees for uncertainty quantification remain scarce. A key obstacle is that the NPMLE is necessarily discrete, which yields discrete posterior credible sets and a slow logarithmic deconvolution rate. We address both limitations by introducing a hierarchical Gaussian smoothing layer that restricts the mixing distribution to a Gaussian location mixture. Our smooth NPMLE inherits the favorable properties of the classical NPMLE: it is computable via convex optimization and achieves nearly parametric denoising performance. Moreover, it achieves a polynomial deconvolution rate that is asymptotically minimax over the corresponding class. Our procedure also leads to estimated smooth posteriors that converge to the true posteriors at a polynomial rate. Further, we characterize marginal coverage sets that are optimal in expected length, construct plug-in estimators of these sets, and establish theoretical guarantees for the estimated sets in terms of both coverage probability and expected length. We also extend the theory to settings with model misspecification and heteroscedastic Gaussian observations, and study identifiability of the proposed hierarchical model.

math.ST↗
arXiv2026

Noise-adaptive hybrid quantum convolutional neural networks based on depth-stratified feature extraction

Hierarchical quantum classifiers, such as quantum convolutional neural networks (QCNNs), represent recent progress toward designing effective and feasible architectures for quantum classification. However, their performance on near-term quantum hardware remains highly sensitive to noise accumulation across circuit depth, calling for strategies beyond circuit-architecture design alone. We propose a noise-adaptive hybrid QCNN that improves classification under noise by exploiting depth-stratified intermediate measurements. Instead of discarding qubits removed during pooling operations, we measure them and use the resulting outcomes as classical features that are jointly processed by a classical neural network. This hybrid hierarchical design enables noise-adaptive inference by integrating quantum intermediate measurements with classical post-processing. Systematic experiments across multiple circuit sizes and noise settings, including hardware-calibrated noise models derived from IBM Quantum backend data, demonstrate more stable convergence, reduced loss variability, and consistently higher classification accuracy compared with standard QCNNs. Moreover, we observe that this performance advantage significantly amplifies as the circuit size increases, confirming that the hybrid architecture mitigates the scaling limitations of standard architectures. Notably, the multi-basis measurement variant attains performance close to the noiseless limit even under realistic noise. While demonstrated for QCNNs, the proposed depth-stratified feature extraction applies more broadly to hierarchical quantum classifiers that progressively discard qubits.

quant-ph↗
arXiv2026

Characterization and cancellation of power-line-induced motional-mode frequency noise in a trapped-ion system

The stability of motional-mode frequency is essential for realizing high-fidelity quantum gates in trapped-ion quantum computing. While broadband Gaussian noise has been extensively studied and mitigated using pulse shaping techniques, the impact of coherent periodic noise has remained largely unexplored. Here we report a systematic investigation of 60-Hz power-line noise and its effect on the secular frequencies of a single ${}^{171}\mathrm{Yb}^{+}$ ion. Using spin-echo Ramsey spectroscopy, we characterize the amplitude and phase of the resulting secular-frequency modulation and validate this characterization via passive phase correction of the Ramsey sequence. Building on this, we implement a cancellation scheme by injecting a compensation tone into the set-point of a PI controller that stabilizes the trap RF drive amplitude. A phasor-fitting procedure optimizes the amplitude and phase of the compensation signal, enabling near-complete suppression of the 60-Hz component. With the cancellation applied, the coherence time of a radial motional mode is extended from approximately 10 ms to 35 ms, consistent with the limit set by motional heating. Our results provide both a clear characterization of periodic motional-mode noise and a practical framework for its suppression in trapped-ion quantum computing platforms.

quant-ph↗
arXiv2026

A rigorous hybridization of variational quantum eigensolver with classical neural network

Combining variational quantum process with classical neural learning offers a flexible route to improve ground-state estimation. We establish an integrated framework that connects neural transformations, measurement statistics, and guarantees physical stability through three requirements: self-contained training, polynomial resource scaling, and variational consistency. Its constructive realization, termed \emph{unitary variational quantum-neural hybrid eigensolver}~(U-VQNHE), couples a variational quantum circuit to a neural phase function through norm-preserving post-processing. The learned transformation is evaluated from measurement records, preserves the exact variational bound, and admits range-independent concentration guarantees for independent finite-shot evaluations. Complementing this construction, we characterize the statistical and representational constraints of amplitude reweighting: sampled-support mismatch can destabilize empirical normalization, while exact distribution matching can require exponentially large dynamic range for Haar-random targets and structured ansatz--target pairs under specified near-tensorizability and mismatch conditions. Finite-shot simulations on Ising and disordered XYZ spin models demonstrate improved energy accuracy over the underlying variational quantum eigensolver and greater robustness than the amplitude-reweighting baselines. Together, these results provide a principled foundation for quantum--neural eigensolvers in which physical consistency, expressive capacity, and measurement cost are treated as a single design problem.

quant-ph↗
arXiv2026

A nuclear disc at Cosmic Noon: evidence of early bar-driven galaxy evolution

Recent studies have revealed that bars can form as early as a few billion years after the Big Bang, already displaying characteristics similar to those of evolved bars in the Local Universe. Bars redistribute angular momentum throughout the galaxy, regulating star formation, AGN activity, and the formation of new stellar structures such as nuclear discs. However, the effects of bar-driven evolution on young galaxies are not yet known, as no evidence of bar-built stellar structures has ever been found beyond $z = 1$, until now. In this work, we present evidence for a bar-built, star-forming nuclear disc already present at redshift $z = 1.5$. This is the first evidence of a bar-built stellar structure at Cosmic Noon. We find that this nuclear disc is actively forming stars and is of similar size to some nuclear discs in nearby galaxies. This evidence solidifies the now emerging picture in which bars are fundamental not only in the late evolution of galaxies, but also in their early evolutionary stages. It changes the current paradigm by urging a revision of our picture of galaxy evolution beyond redshift one to include new considerations of the role of bars as early as a few billion years after the Big Bang.

astro-ph.GA↗
arXiv2026

Multimode Phonon-Number Measurement and Single-shot Superparity Measurement using Dispersive Shifts in a Trapped Ion

Dispersive shifts are a widely used tool for bosonic readout and control in circuit quantum electrodynamics, yet they remain relatively unexplored in trapped-ion motional systems. Here we introduce a unified framework for multimode phonon-number measurement and nondestructive single-shot superparity measurement, i.e., phonon-number measurement modulo 2^k, using dispersive shifts in the far-detuned multimode Jaynes-Cummings interaction of a trapped ion system. We implement a Ramsey sequence that realizes a multimode spin-dependent rotation (SDR) together with a selective decoupling scheme that cancels the phase induced by the carrier AC-Stark shift while preserving the phonon-number-dependent phase induced by the dispersive shift. Within this framework, we infer single-mode and two-mode Fock-state distributions from spin-population dynamics, use SDR-based conditional parity operators with postselection to generate cat states and entangled coherent states, and realize nondestructive single-shot measurements of phonon number modulo 2, 4, and 8 in the single-mode setting. These results open a new avenue for the use of multimode parity operators in trapped-ion bosonic systems.

quant-ph↗
arXiv2025

The impact of bars on the properties of HII regions in the TIMER survey

In this study we perform a comparative analysis of the properties of the HII regions located in different areas of barred galaxies, with the aim of investigating the impact of bars on the physical properties of the ionised gas. Based on integral field spectroscopy data for 17 barred galaxies covering approximately the central 6x6 kpc, we detect a total of 2200 HII regions, of which 331 are located within the nuclear disc (also known as circumnuclear regions), 661 in the bar region, and 1208 in the disc. Among the physical properties of the HII regions, we explore the O/H and N/O abundances, H$\alpha$ luminosity, dust extinction, electron density, and H$\alpha$ equivalent width. We find clear differences in the properties of the HII regions between the nuclear disc, the bar and the disc, that could be explained by an enhancement in the molecular gas concentration in the central parts driven by bar-induced gas flows. As this gas is channelled towards the galaxy centre, the most extreme values in the analysed properties are found for the circumnuclear HII regions. Unlike the bar strength, galaxy mass does seem to affect the properties of the HII regions, with massive galaxies presenting higher values in most of the properties, possibly due to the increased amount of gas in these systems. This study provides evidence that the bar-driven redistribution of material within the galaxy inner parts causes significant differences in the HII region properties depending on their location within the galaxies.

astro-ph.GA↗
arXiv2025

Motif-2-12.7B-Reasoning: A Practitioner's Guide to RL Training Recipes

We introduce Motif-2-12.7B-Reasoning, a 12.7B parameter language model designed to bridge the gap between open-weight systems and proprietary frontier models in complex reasoning and long-context understanding. Addressing the common challenges of model collapse and training instability in reasoning adaptation, we propose a comprehensive, reproducible training recipe spanning system, data, and algorithmic optimizations. Our approach combines memory-efficient infrastructure for 64K-token contexts using hybrid parallelism and kernel-level optimizations with a two-stage Supervised Fine-Tuning (SFT) curriculum that mitigates distribution mismatch through verified, aligned synthetic data. Furthermore, we detail a robust Reinforcement Learning Fine-Tuning (RLFT) pipeline that stabilizes training via difficulty-aware data filtering and mixed-policy trajectory reuse. Empirical results demonstrate that Motif-2-12.7B-Reasoning achieves performance comparable to models with significantly larger parameter counts across mathematics, coding, and agentic benchmarks, offering the community a competitive open model and a practical blueprint for scaling reasoning capabilities under realistic compute constraints.

cs.AI↗
arXiv2025

Motif 2 12.7B technical report

We introduce Motif-2-12.7B, a new open-weight foundation model that pushes the efficiency frontier of large language models by combining architectural innovation with system-level optimization. Designed for scalable language understanding and robust instruction generalization under constrained compute budgets, Motif-2-12.7B builds upon Motif-2.6B with the integration of Grouped Differential Attention (GDA), which improves representational efficiency by disentangling signal and noise-control attention pathways. The model is pre-trained on 5.5 trillion tokens spanning diverse linguistic, mathematical, scientific, and programming domains using a curriculum-driven data scheduler that gradually changes the data composition ratio. The training system leverages the MuonClip optimizer alongside custom high-performance kernels, including fused PolyNorm activations and the Parallel Muon algorithm, yielding significant throughput and memory efficiency gains in large-scale distributed environments. Post-training employs a three-stage supervised fine-tuning pipeline that successively enhances general instruction adherence, compositional understanding, and linguistic precision. Motif-2-12.7B demonstrates competitive performance across diverse benchmarks, showing that thoughtful architectural scaling and optimized training design can rival the capabilities of much larger models.

cs.CL↗
arXiv2025

Dark gaps and resonances in barred galaxies

Dark gaps, low surface brightness regions along the bar minor axis, are expected to form as a consequence of secular evolution in barred galaxies. Although several studies have proposed links between dark gap locations and dynamical resonances, the results remain inconclusive. Using DESI Legacy Imaging Survey data, we find that approximately 61% of barred galaxies exhibit pronounced dark gaps. We compare the location of dark gaps with resonance radii derived from the Tremaine-Weinberg method applied to MaNGA data for the same galaxies. Our analysis shows that dark gaps do not preferentially form at specific resonances. Instead, their locations correlate with $\mathcal{R}$ $\equiv$ $R_{CR}/R_{Bar}$: slow bars tend to show shorter dark gap radii, while fast bars show longer ones. This trend reflects a tight relation between bar length and dark gap radius. However, when barred galaxies are classified by their ring morphology, certain types exhibit dark gaps that align with specific resonances. Notably, dark gaps located between the inner and outer rings are closely associated with the corotation radius. In galaxies with two dark gaps along the bar minor axis profile, the inner dark gap typically aligns with the ultraharmonic resonance, and the outer dark gap corresponds to the corotation radius. These findings suggest that some morphological types share similar $\mathcal{R}$ values and exhibit dark gaps near specific resonances. Thus, dark gaps may serve as proxies for dynamical resonances only in certain systems. Our findings may help explain the discrepancies observed in earlier studies.

astro-ph.GA↗