arXiv subjects

Kun Huang

Publications and source records attributed to Kun Huang.

At least 19 recordsLinked to original sources

arXiv2026

GSAR: Goal-State-Anchor Rewards for Mobile GUI Agents with Self-Evolving Data Synthesis

Vision-Language Models (VLMs) based GUI agents stand to benefit significantly from online reinforcement learning (RL). However, their training is bottlenecked by two fundamental issues: current data synthesis methods for GUI Agents rely on specific environments and struggle to generate diverse data, while existing evaluators either suffer from limited scalability or provide inaccurate and unreliable reward signals. To overcome these challenges, we introduce GSAR (Goal-State-Anchor Reward), a RL reward framework that supports scalable task generation and delivers reliable reward signals for stable and efficient policy optimization. Our approach features self-evolving data synthesis, which produces multiple environments through task execution and generates diverse tasks and goal states. Complementing this, a state-anchor mechanism automatically annotates task-relevant UI elements in successful goal states as reference anchors. During RL training, these reference anchors provide accurate, scalable reward signals that substantially enhance efficiency. Extensive evaluations demonstrate that our framework achieves over 90% accuracy on offline trajectory verification and performs closest to rule-based methods. Furthermore, agents trained using our reward framework exhibit strong performance on both AndroidWorld and our constructed benchmark, establishing a scalable approach for GUI agent training.

cs.AI↗

arXiv2026

Weak Identification in Bunching Designs: Rare Events and Counterfactual Uncertainty

We study shrinking tax kinks when the counterfactual density is locally Hölder continuous and may be accompanied by a known background atom at the threshold. Small policy incentives create two distinct obstacles: few behavioral bunchers and an unobserved relocation interval. For the complete observed distribution, we establish matching minimax bounds for elasticity estimation and honest confidence-set diameter. With kink size $d$, background mass $γ$, and smoothness $0<s\leq1$, the rate is $\min\{1,\sqrt{(γ+d)/(nd^2)}+d^s\}$. The stochastic component has a rare-event scale when the background mass is of the kink order or smaller and a conventional weak-signal scale when it remains positive. The extrapolation component follows from observationally equivalent counterfactual densities, rather than sampling error. We derive normal and Poisson limits for a conventional bunching ratio, show a coverage failure of its Wald interval, and give finite-sample confidence sets by exact binomial inversion with a deterministic smoothness allowance. The procedure estimates the counterfactual density from an unaffected bin and covers the entire population identified set uniformly, including zero policy incentives. A likelihood-ratio calculation establishes a finite-parameter Poisson limit for a smooth, known counterfactual submodel. Structural Monte Carlo experiments document both coverage protection and its cost in set length. The results apply to the stated exact-atom observation model; general rounding and optimization frictions require additional assumptions.

econ.EM↗

arXiv2026

Existence and Smoothing for a Nondivergence-Form Degenerate Diffusion from Plasma-Wave Theory

We prove existence and positive-time smoothing of weak solutions to the degenerate parabolic Cauchy--Dirichlet problem $\partial_t u = ρ_λ(x)u\,\partial_x^2 u + ρ_λ(x)g(x)u$ on the half-line, where $ρ_λ$ vanishes at the boundary and grows at infinity. This scalar problem arises from a formal reduction of the one-dimensional quasilinear system for plasma waves, which couples an electron distribution to a wave-energy spectrum: we develop a solution theory for the scalar problem that lifts to this system. For the scalar problem, we construct nonnegative weak solutions under weighted integrability assumptions on the initial data and boundedness of the reaction coefficient $ρ_λg$, admitting unbounded and discontinuous data pertinent to the physical model. We establish a one-sided bound on $\partial_t u$ which yields positive-time smoothing: solutions become locally Hölder continuous in both space and time and locally Lipschitz continuous in space. Explicit examples show that this spatial regularity is sharp. Under natural assumptions on the initial data, our scalar solutions lift to weak solutions of the particle--wave system that preserve nonnegativity, conserve particle mass, smooth initial particle measures to locally bounded densities at positive times, and exhibit quantitative decay of the wave spectrum at large wavenumber.

math.AP↗

arXiv2026

Weak solutions for weak turbulence models in electrostatic plasmas

The weak turbulence model, also known as the quasilinear theory in plasma physics, has been a cornerstone in modeling resonant particle-wave interactions in plasmas. Stemming from the Vlasov-Poisson/Maxwell system under the weak turbulence assumption, incorporating the random phase approximation and ergodicity, it takes the form of a integro-differential system, which describes the time dynamics of particle distribution and wave spectrum. The interaction between particles and waves can be treated as a stochastic process, whose transition probability bridges the momentum space and the spectral space. To our knowledge, there is no preceding rigorous analytical work for the quasilinear theory addressed in this manuscript. In this paper, we establish the existence of global weak solutions to the one-dimensional electrostatic quasilinear model for nonnegative, smooth, rapidly decaying initial particle distributions and nonnegative, smooth wave spectra compactly supported in a positive wave-number band. The construction uses the resonance condition to reduce the coupled system to a scalar degenerate parabolic equation, together with uniform estimates and strong convergence of the gradients of regularized approximations. When the initial wave spectrum is positive throughout a single band, we further characterize the equilibrium selected by this approximation procedure. The equilibrium uniquely minimizes a quadratic functional over a nonnegative cone and admits a geometric construction through the least concave majorant of a primitive determined by the initial data. The limiting particle distribution is non-increasing on the corresponding resonant interval and forms plateaus wherever the equilibrium auxiliary function is positive. A relative-entropy argument establishes long-time convergence to this equilibrium.

math.AP↗

arXiv2026

Distributed Stochastic Momentum Tracking with Local Updates: Achieving Optimal Communication Complexities

We propose Local Momentum Tracking (LMT), a novel distributed stochastic gradient method for solving distributed optimization problems over networks. To reduce communication overhead, LMT enables each agent to perform multiple local updates between consecutive communication rounds. Specifically, LMT integrates local updates with the momentum tracking strategy and the Loopless Chebyshev Acceleration (LCA) technique. We demonstrate that LMT achieves linear speedup with respect to the number of local updates as well as the number of agents for minimizing smooth objective functions with and without the Polyak-Łojasiewicz (PL) condition. In particular, LMT outperforms existing distributed stochastic gradient methods with local updates, and notably attains the optimal communication complexities with sufficiently many local updates $Q\geq Q^*$. For a moderate number of local updates $Q=\mathcal{O}(1)$, the performance of LMT matches the best known result for $Q=1$ under smooth nonconvex objectives, and improves upon it under the PL condition. For $Q=1$, we further establish minimax performance lower bounds over two comparison families. For smooth objectives, LMT matches the lower bound. For PL objectives, LMT attains the minimax-optimal dependence on the spectral gap, up to logarithmic factors. To our knowledge, LMT is the first distributed stochastic gradient method with local updates that enjoys such properties.

math.OC↗

arXiv2026

Nuclear mass table in deformed relativistic Hartree-Bogoliubov theory in continuum, III: nuclei with $8 \leq Z \leq 120$

The mass table in the deformed relativistic Hartree-Bogoliubov theory in continuum (DRHBc) with the PC-PK1 density functional has been established for nuclei with $8 \leq Z \leq 120$, extended from the previous works for even-even nuclei [Zhang et al. (DRHBc mass table collaboration), At. Data Nucl. Data Tables 144, 101488 (2022)] and for even-$Z$ nuclei [Guo et al. (DRHBc mass table collaboration), At. Data Nucl. Data Tables 158, 101661 (2024)]. The calculated binding energies, two- and one-nucleon separation energies, root-mean-square (rms) radii of neutron, proton, matter, and charge distributions, quadrupole deformations, neutron and proton Fermi surfaces, and the blocked neutron (proton) orbitals of odd-$N$ ($Z$) nuclei are tabulated and compared with the available experimental data. A total of 9495 nuclei are predicted to be bound, with an rms deviation of 1.444 MeV from the 2380 mass data. Good agreement with the available experimental pairing gaps, $α$ decay energies, and charge radii is also achieved. The accuracies of the calculated nuclear masses and nucleon separation energies as well as the prediction for drip lines are compared with those obtained by other relativistic and nonrelativistic density functional calculations. It turns out that the DRHBc theory with PC-PK1 provides one of the best microscopic descriptions for nuclear masses. The systematics of nucleon separation energies, pairing gaps, pairing energies, two-nucleon gaps, $α$ decay energies, rms radii, quadrupole deformations, potential energy curves, neutron density distributions, and neutron mean-field potentials are discussed.

nucl-th↗

arXiv2026

HOCCL: Offloading Collective Communication from GPU Cores to Accelerate Distributed Training

Large language model training involves massive computation on GPU streaming multiprocessors (SMs), the primary compute units of GPUs. Since SMs host specialized accelerators such as Tensor Cores, their efficient utilization is critical to training efficiency. Unfortunately, existing collective communication systems compete with computation for SMs, as they consume SMs for communication-related data movement and synchronization operations. We observe that communication can, in principle, be driven by DMA engines, thereby eliminating SM involvement in communication. Based on this insight, we propose HOCCL, a zero-SM collective communication framework consisting of three components: a stream manager, a point-to-point (P2P) executor, and a collective scheduler. The stream manager preserves operator-level temporal ordering with other GPU kernels. The P2P executor enables zero-SM point-to-point communication, while the collective scheduler orchestrates P2P transfers to maximize bandwidth. Experiments show that HOCCL preserves near-peak communication performance, achieving within 3% of the state of the art on average, while eliminating communication occupancy on nearly 10% of total GPU SMs. By freeing SM resources for computation, HOCCL improves end-to-end training throughput by up to 5%.

cs.NI↗

arXiv2026

Where to Look and What to Use: Retrieve-Localize-Generate for Long-Term Conversational Memory Question Answering

Retrieval-augmented generation (RAG) enables large language models (LLMs) to answer questions by accessing external knowledge and has been widely adopted for long-term conversational memory question answering. However, existing methods suffer from two key challenges: (1) fragmented evidence scattered across temporally distant sessions, and (2) noisy content within retrieved sessions that triggers the lost-in-the-middle effect. To address these challenges, we propose MemLoc, a unified Retrieve-Localize-Generate framework for long-term conversational memory QA. For retrieval, MemLoc decomposes each session into multi-granularity memory units and performs query routing via an inner-memory graph with entropy-based granularity selection. It further models cross-session semantic and temporal dependencies through a cross-memory graph, enabling coarse-to-fine retrieval of top-K relevant memory candidates. For localization, we introduce a reasoning-based evidence locator trained with Self-reflective Hint Policy Optimization (SHPO), which performs progressive refinement by extracting query-relevant fragments within memory units to suppress noise and reranking across candidates to remove redundancy, producing a compact evidence set with lightweight location IDs. For generation, these IDs act as precise grounding signals that guide the LLM to the correct memory positions, mitigating the lost-in-the-middle effect while preserving original contextual integrity. Extensive experiments on four benchmarks demonstrate that MemLoc achieves state-of-the-art retrieval accuracy and response quality while maintaining efficiency. Our code is available at: https://github.com/Nikol-coder/MemLoc.

cs.CL↗

arXiv2026

Consistent Bayesian Spatial Domain Partitioning Using Predictive Spanning Tree Methods

Bayesian model-based spatial clustering methods are widely used for their flexibility in estimating latent clusters with an unknown number of clusters while accounting for spatial proximity. Many existing methods are designed for clustering finite spatial units, limiting their ability to make predictions, or may impose restrictive geometric constraints on the shapes of subregions. Furthermore, the posterior clustering consistency theory of spatial clustering models remains largely unexplored in the literature. In this study, we propose a Spatial Domain Random Partition Model (Spat-RPM) and demonstrate its application for spatially clustered regression, which extends spanning tree-based Bayesian spatial clustering by partitioning the spatial domain into disjoint blocks and using spanning tree cuts to induce contiguous domain partitions. Under an infill-domain asymptotic framework, we introduce a new distance metric to study the posterior concentration of domain partitions. We show that Spat-RPM achieves a consistent estimation of domain partitions, including the number of clusters (which may go to infinity), and derive posterior concentration rates for partition, parameter, and prediction. We also establish conditions on the hyperparameters to achieve consistency, offering important practical guidance for hyperparameter selection. Finally, we examine the asymptotic properties of our model through simulation studies and apply it to Atlantic Ocean data.

stat.ME↗

arXiv2026

Wasserstein Stability and Free Boundaries in Measure-Parameterized Bilevel Obstacle Problems

We study obstacle-constrained variational problems whose reduced energies depend on a probability law through a lower-level optimizer. Uniform strong convexity yields a single-valued Lipschitz follower response, while convexity and a Poincare inequality give a unique upper-level policy. A type-Lipschitz reduced marginal then implies Lipschitz continuity of the policy map from the 1-Wasserstein metric to the energy space. In a one-dimensional linear-obstacle subclass, the policy derivative is the positive part of a cumulative forcing. Single crossing makes the coincidence set an interval. An algebraic crossing of order m gives a $W_1^{1/m}$ modulus for its endpoint; odd-power examples show that this exponent is sharp, while a transversal crossing recovers Lipschitz stability. For empirical laws on compact subsets of $\mathbb{R}^k$, dimension-dependent Wasserstein bounds yield finite-sample rates for policies and thresholds. At a transversal population root, the empirical threshold is asymptotically linear, with an explicit influence function and central limit theorem. For higher-dimensional regular patches, a conditional level-set argument gives local Hausdorff stability when a nondegenerate switching function is available. An explicit quadratic firm response produces a nonlinear corporate-tax schedule with an endogenous zero-tax region and a Wasserstein-stable threshold. The tax illustration is analytic and uses no empirical calibration.

math.OC↗

arXiv2026

Blockchain-based Proportional Fair Scheduling for Multi-Operator O-RAN

The openness and disaggregation of Open radio access network (O-RAN) facilitate resource sharing and coordination across networks, creating new demands for efficient and trustworthy cross-operator scheduling. However, such scheduling is beyond the scope and capability of conventional proportional fair scheduling (PFS), which lacks mechanisms for establishing trust among independent operators. To fulfill this gap, we propose the blockchain-based proportional fair scheduling (BC-PFS) that enables trustworthy inter-network coordination and resource pooling across operators in O-RAN. Specifically, we design four core smart contracts including registration, status reporting, scheduling, and settlement contracts with corresponding Solidity implementations to ensure trustworthy on-chain execution. Theoretically, to evaluate the BC-PFS performance, we develop an analytical framework to derive the user average throughput via both probabilistic and ordinary differential equation (ODE) approaches, and provide a simplified closed-form solution. Based on the above performance assessment, we quantify the pooling effect in O-RAN achieved through trustworthy cross-operator collaboration via BC-PFS, and point out that this effect grows monotonically in both the numbers of operator networks and users. Simulations validate the theoretical analysis and show the performance of the BC-PFS in O-RAN.

cs.NI↗

arXiv2026

Runaway electron control by self-excited waves

Runaway-electron avalanches in tokamak plasmas can be limited by kinetic instabilities driven by the non-Maxwellian runaway distribution. We formulate a reduced model for the quasi-steady state in which the total plasma current and bulk electron temperature are prescribed, while the inductive electric field is determined self-consistently from the partition between Ohmic bulk current and runaway-electron current. Because the wave growth time is short compared with the current-decay time, we consider a marginal-stability regime, in which whistler-wave drive by the runaway electrons balances collisional damping. The resulting states separate into three regimes: a subcritical Ohmic regime without an avalanche, an avalanche regime in which runaway growth relaxes the inductive field to the avalanche threshold, and an instability-regulated regime in which self-excited whistler waves enhance momentum-space diffusion and limit the runaway current. In the instability-regulated regime, the whistler wave spectrum forms a narrow ridge, and low-energy runaway electrons carry most of the runaway current.

physics.plasm-ph↗

arXiv2026

Distributed risk-averse optimization via CVaR

Distributed systems often operate under uncertainty, where minimizing expected loss may overlook rare but severe events. This paper studies a distributed risk-averse convex optimization problem in which agents cooperatively minimize the average of local conditional value-at-risk (CVaR) objectives over a time-varying network. Each agent has access only to noisy evaluations of its local loss function, rather than to its CVaR objective or gradient. We therefore develop a zeroth-order algorithm that uses sampled losses to construct empirical CVaR estimates and their gradient estimates. At each iteration, agents combine neighboring decisions and perform a local update. Under convexity and Lipschitz continuity assumptions, we prove that the agents reach exact asymptotic consensus. We also establish a finite-time expected suboptimality bound for the weighted ergodic iterate. With diminishing step sizes and fixed sample sizes, the local last iterates converge almost surely to a common optimum, and their limiting expected CVaR gap is bounded in terms of the smoothing and finite-sample errors. This distributed bound matches the parameter dependence of the centralized benchmark provided in this paper. Finally, simulations on a distributed sensor network estimation problem illustrate the efficacy of the method.

math.OC↗

arXiv2026

AppDeltaWorld: Transition-Grounded Delta Code World Model for Mobile GUI Agents

Mobile GUI agents can operate apps through pixel perception and touch actions, making them a promising interface for collecting and improving long-horizon mobile interaction policies. However, real trajectories are difficult to obtain for sensitive apps and privacy-critical operations. At the same time, existing simulated environments are costly to scale up, and GUI world models still suffer from unstable generation, limited modality coverage, and inconsistent action-transition logic. To address these limitations, we propose AppDeltaWorld, a transition-grounded delta code world model that predicts the next GUI as a reachable code update rather than as an unconstrained image or text description. AppDeltaWorld retrieves app-specific Level-1 HTML references under an action-transition constraint, generates Level-2 executable HTML conditioned on the current screen, action, predicted next-screen text, and retrieved structure, and inserts generated visual assets into image slots before browser rendering. As a world model, AppDeltaWorld achieves the highest fidelity on CMGUIBench-500 under Code2World evaluation, with clear gains in structural layout and UI element reconstruction over image-only and code-only baselines. As a training environment, AppDeltaWorld supports filtered closed-loop SFT data construction that, when combined with public supervision, enables AppDeltaAgent to achieve state-of-the-art performance on AndroidLens and consistent gains on MobileGym and MobileWorld. Moreover, world-model-based test-time reinforcement learning enables policy adaptation and shows further improvements without additional interaction with real apps.

cs.AI↗

arXiv2026

A fast scheme for the homogeneous Boltzmann equation based on lifting and tensor train approximation

We propose a fast deterministic scheme for the space-homogeneous Boltzmann equation that exploits the low-rank structure of the velocity distribution. This paper consists of two independent contributions. The first is a \emph{lifting-projection (LP) scheme}, inspired by the approach in the recent theoretical breakthroughs \cite{guillen2025landau, imbert2026monotonicity, guillen2025landau2} on the well-posedness of the Landau and Boltzmann equations. In particular, the approach lifts the nonlinear 3D Boltzmann equation to the 6D linear Kac master equation, advanced over a single time step, and projected back to its marginal in 3D. The second contribution is a \emph{low-rank tensor method} for evaluating the collision operator, in which the lifted solution is represented in tensor train (TT) format and computed via a TT cross approximation algorithm with interpolation, complemented by a TT-friendly conservation correction that enforces conservation of mass, momentum, and energy. When the solution is low-rank in velocity, the method scales linearly in $n$ when cubic interpolation is used (and quadratic in $n$ when spectral interpolation is used), where $n$ is the number of grid points in each velocity direction. Therefore, our methods offer significant computational savings over existing deterministic solvers in such cases. Numerical experiments on 2D and 3D benchmarks, including the BKW exact solution and anisotropic initial data, confirm the computational scaling, the expected order of accuracy and verify the effectiveness of the conservation correction.

math.NA↗

arXiv2026

Reconstructive comb spectroscopy: A single-pixel detection paradigm beyond dual-comb limitations

Frequency comb spectroscopy has revolutionized broadband molecular fingerprinting with mode-defined resolution. While dual-comb spectroscopy stands as a dominant paradigm for high-resolution measurements, it relies on mutually coherent dual combs, and its applicability to non-cooperative sensing is limited by the requirement for phase-sensitive detection and controlled optical returns. Here, we introduce reconstructive comb spectroscopy, a fundamentally different paradigm that eliminates these constraints. By integrating a mode-programmable optical comb with a computational sensing scheme based on single-pixel detection, our method achieves picometer-level spectral resolution over a 10-nm (1.27-THz) instantaneous bandwidth, with single-photon sensitivity down to 10^-4 photons per pulse, and compressed spectral acquisition at 2.5% sampling while maintaining reconstruction errors below 10%. We demonstrate robust performance through scattering media and from non-cooperative targets. These capabilities establish reconstructive comb spectroscopy as a new platform for gas sensing, with broad applicability in remote atmospheric monitoring, industrial leak detection, and standoff chemical-threat identification.

physics.optics↗

arXiv2026

Structure-Aware Compound-Protein Affinity Prediction via Graph Neural Networks with Group Lasso Regularization

Explainable artificial intelligence approaches accelerate drug discovery by improving molecular representation learning, identifying key molecular structures, and rationalizing drug property prediction. However, developing end-to-end explainable models for target-specific structure-activity relationship modeling remains challenging because compound-protein interaction data are often limited for individual targets, and small changes in chemical substituents or local structural motifs can cause large differences in molecular properties. Therefore, effectively leveraging structural and property information to identify key moieties associated with compound-protein affinity is essential. We propose a graph neural network (GNN) framework that uses property and structural information from activity-cliff molecule pairs targeting specific proteins to predict compound-protein affinity, measured by half-maximal inhibitory concentration (IC50), and explain property differences. To improve explainability, we trained GNNs with structure-aware loss functions using group lasso and sparse group lasso regularization, which prune and highlight molecular subgraphs relevant to activity differences. We applied this framework to activity-cliff data from molecules targeting six tyrosine-protein kinases across the Src, Abl, and Tec families, as well as anaplastic lymphoma kinase. Integrating common- and uncommon-node information with sparse group lasso improved target-specific molecular property prediction, producing lower root mean square errors and higher Pearson correlation coefficients. Regularization also enhanced GNN feature attribution by improving graph-level global direction scores and atom-level coloring accuracy. These results support more interpretable drug discovery pipelines, particularly for identifying critical molecular substructures during lead optimization.

cs.LG↗

arXiv2026

Single-photon time-stretch computational ghost spectroscopy

Time-stretch spectroscopy is powerful for capturing transient spectral phenomena but remains fundamentally limited by detector bandwidth or timing jitter, especially under photon-starved conditions. Here, we devise and implement single-photon time-stretch computational ghost spectroscopy, which integrates dispersive wavelength-to-time mapping with programmable temporal encoding and correlation-based reconstruction to overcome these detection limitations. Specifically, temporally stretched ultrashort pulses are modulated by predefined encoding patterns and detected by a low-bandwidth detector, allowing reconstruction of near-infrared spectra with 450 resolvable channels across 1530-1590 nm without direct high-speed waveform acquisition. By further incorporating compressive sensing, accurate spectral recovery is achieved at sub-Nyquist sampling rates, substantially reducing acquisition requirements to facilitate high-speed operation at 210 kHz. In the single-photon regime, computational ghost reconstruction effectively suppresses the intrinsic detector timing jitter, yielding high-fidelity spectra at illumination fluxes down to 0.01 photons/pulse. By jointly enabling broadband coverage, high spectral resolution, high acquisition speed, and single-photon sensitivity, this approach establishes a computation-enhanced paradigm for time-stretch spectroscopy and provides a versatile platform for ultrafast and photon-efficient spectroscopic applications.

physics.optics↗