arXiv subjects

Lin Wang

Publications and source records attributed to Lin Wang.

At least 19 recordsLinked to original sources

arXiv2026

On the Dynamics of Invariant Graphs for Dissipative Twist Maps

We study the persistence and internal dynamics of invariant graphs for dissipative twist maps. Explicit $C^1$ persistence and parameter estimates allow every real rotation number to be realized by varying either translation parameter, while arbitrarily small zero-mean analytic perturbations at fixed irrational horizontal parameters can yield analytic graphs with dynamics conjugate to rational rotations. With a fixed Brjuno horizontal parameter and fixed-strip analyticity, normally attracting $C^r$ graphs persist under perturbations of size $(1-λ)^γ$, for any $γ>0$ and prescribed finite $r$, as $λ\to1^-$. At the critical Lipschitz bound, the invariant graphs can carry Denjoy dynamics and have countably many nondifferentiable points, with perturbations smooth outside a null Cantor set and tending to zero in every Hölder norm of order less than one.

math.DS↗

arXiv2026

UAV-DualCog: A Dual-Cognition Benchmark for Grounded UAV Agent-World Reasoning with MLLMs

Multimodal large language models (MLLMs) perform strongly across vision-language tasks, yet remain insufficiently explored in UAV scenarios. Recent UAV-oriented benchmarks have begun to evaluate MLLMs in aerial scenarios, but typically focus on scene understanding, event recognition, or navigation completion, rather than jointly assessing the dual-cognition capability required for UAV agents: reasoning about the UAV's own position, behavior, and observation state alongside external landmark states under continuously changing viewpoints, supported by explicit spatio-temporal grounding. To address this gap, we present UAV-DualCog, the first benchmark to formulate UAV dual cognition as the coordinated evaluation of Agent-State Reasoning and World-State Reasoning across multi-view images and egocentric videos, with explicit spatial or temporal grounding. The State-aware UAV Simulation Platform (SUSP) and the DualCog-aware Capture System (DCCS) construct 4,096 image and 2,048 video samples from 12 scenes, pairing complementary state questions within shared Agent-Landmark contexts. Extensive evaluations on 45 models show that current MLLMs remain far from reliable in UAV dual cognition: Agent-State Reasoning is generally weaker than World-State Reasoning, semantic answers do not reliably coincide with spatial or temporal grounding, and coordinated abilities remain limited. Thinking-mode comparisons, open-ended analyses, and sim-to-real validation show that these limitations persist across inference settings, formats, and environments. We further construct UAV-DualCog-Train and improve UAV dual-cognition performance through a lightweight optimization probe. Supplementary website, code, and dataset: https://smartdianlab.github.io/UAV-DualCog/

cs.CV↗

arXiv2026

Full edge control of bulk energy gap in armchair graphene nanoribbons

In solid-state physics, the ability to open and tune an electronic energy gap provides a direct route to controlling charge transport and accessing distinct electronic phases. In solids with physical boundaries, a bulk energy gap formed exclusively by extended bulk states is conventionally expected to be governed primarily by the interior bulk region, with only a minor contribution from the boundaries. Here we show that, in stark contrast to this conventional expectation, the bulk gap of armchair graphene nanoribbons in the $N_w=3p+2$ family with $p\in\mathbb{Z}\ge 0$ can be fully controlled through the hoppings solely at the boundaries. This full boundary control originates from an exact cancellation between the gap contributions of long-range hoppings along and across the ribbon in the bulk region. It persists in the presence of Hubbard interactions and holds irrespective of ribbon width in this family. Our findings establish boundary dynamics as an unexpectedly powerful means of manipulating bulk electronic properties, enabling width-preserving engineering of one-dimensional (1D) band structures and topological phases, with potential applications in qubits, artificial quantum spin chains, topological quantum dots and other quantum devices.

cond-mat.mes-hall↗

arXiv2026

Morphological Decoupling-Based Skeletal Classification for Clinical Assessment of Malocclusion

Malocclusion skeletal grading is a fundamental task in orthodontics, critical for diagnosis and treatment planning. Traditionally, cone-beam computed tomography (CBCT) is used for visual measurement, and the reconstructed lateral cephalograms are handed over to expert dentists for diagnosis. However, manual review is time-consuming, labor-intensive, and subject to inter-operator variability. Therefore, an automatic CBCT-based system is needed for reliable malocclusion skeletal grading. In this case, we develop TeethGNN, a novel graph-based framework designed to combine CBCT image features with morphological information for accurate and efficient malocclusion grading. TeethGNN utilizes a decoupled learnable decoder to directly predict key morphological indicators from CBCT images, eliminating the need for manual measurements. These morphological features are then fused with image features using a graph neural network (GNN), which effectively models the relationships between the modalities. To further enhance robustness and calibration, we introduce a collaborative calibration strategy. This strategy combines multi-scale graph adversarial perturbation for explicit calibration and nonlinear topological graph calibration for implicit confidence adjustment. Extensive experiments and ablation studies on our collected clinical dataset demonstrate that our malocclusion measurement system achieves 77.08\% in accuracy and 89.61\% in AUC, outperforming the compared state-of-the-art methods. These results validate the effectiveness of graph-based multimodal fusion and collaborative calibration in improving malocclusion grading performance. Our system shows strong potential for advancing computer-aided orthodontic diagnosis, providing an accurate and reliable solution for vision-based clinical measurement and diagnosis.

eess.IV↗

arXiv2026

Quantitative Persistence and Regularity of Invariant Graphs for Dissipative Twist Maps

In this paper we study the robustness of invariant Lagrangian graphs for dissipative twist maps on $\mathbb T^d\times\mathbb R^d$ in the $C^r$ category, for any $r\ge 1$, and show their persistence as invariant global attractors under quantitative sharp conditions involving both the size of dissipation parameter and the perturbation at stake. Note that when $d>1$ the geometric Lagrangian property of the invariant manifold may be destroyed, so that the graph may persist but fail to be Lagrangian. In general, additional dynamical requirements on the twist maps come into play if we intend to preserve also the Lagrangian property. In this line, we first study their persistence for general perturbations, then we restrict to the class of exact conformally symplectic twist maps and provide quantitative sharp conditions for the persistence of invariant exact Lagrangian graphs, without assuming any condition on the tangential dynamics of the manifold. The persistence result relies on a combination of a suitable multidimensional Herman--Mather formula proved in this setting and contraction estimates that provide sharp bounds on the $C^1$-norm of the perturbation to get a unique, globally attracting $C^r$ graph. In dimension $d=1$, by further assuming that that the circle map induced by the graph has irrational rotation number, we prove that every $C^1$ invariant graph of a $C^k$ map is $C^k$ for $k=2,3,\ldots,\infty$. Analytic maps with Lipschitz, non-$C^1$ invariant graphs at every prescribed irrational rotation number show that this initial regularity assumption is essential.

math.DS↗

arXiv2026

Hybrid Monte Carlo for the extended Kane-Mele-Hubbard model

The Hubbard model offers a powerful minimal description of strongly correlated electrons in molecules and materials, yet it neglects effects observed in nature that give rise to a much richer variety of physics. For instance, including nonlocal interactions introduces a competition between strongly correlated phases, while incorporating spin-orbit coupling (SOC) can lead to the emergence of exotic topological phases. Seeking a more realistic platform for simulating graphene and, potentially, related materials such as transition metal dichalcogenides (TMDs), we therefore consider the extended Kane-Mele-Hubbard model, which features a Kane-Mele SOC term and nearest-neighbor interactions. Simulating this model is particularly challenging once the nonlocal interactions become dominant over the on-site Hubbard interaction, as a severe sign problem arises already at half filling, leaving this regime largely unexplored. In this work, we develop continuous auxiliary field formulations of the model, suitable for treatment with the Hybrid Monte Carlo (HMC) algorithm. Using these formulations, we characterize the sign problem across the model's parameter space and establish where simulations are feasible, as a first step toward the full phase diagram.

cond-mat.str-el↗

arXiv2026

A Large-Volume GAGG(Ce) Detector Module for In-Beam $γ$-Ray Spectroscopy

We report on the development and characterization of large-volume GAGG(Ce) scintillation detector modules equipped with silicon photomultiplier (SiPM) readout. Each module houses four large-volume GAGG(Ce) crystals within a compact thin carbon-fiber enclosure, minimizing the passive material budget and enabling the construction of high-efficiency, large-solid-angle detector arrays for in-beam $γ$-ray spectroscopy. The performance of 10 GAGG modules (40 detectors), including the energy linearity, energy and timing resolution, and intrinsic full-energy peak efficiency, was evaluated using standard $γ$-ray sources and $^{14}$N(p,$γ$)$^{15}$O reactions, covering a $γ$-ray energy range from about 0.3 MeV to 8 MeV. Non-linear energy response across the measured energy range was observed due to the non-linear response of the SiPM. The detector modules exhibit good energy resolution (FWHM below 5\% at 1.33 MeV), good timing resolution (FWHM $\sim$500 ps at $\sim$1 MeV), and intrinsic full-energy peak efficiencies in good agreement with Geant4 simulations, making the detector modules a promising solution for next-generation high efficiency in-beam $γ$-ray experiments at RI-beam facilities.

physics.ins-det↗

arXiv2026

On $C^2$ circle diffeomorphisms with rotation numbers of constant type

For every irrational rotation number $α$ of constant type and every $q>1$, we construct a $C^2$ circle diffeomorphism $F$ arbitrarily close to the rigid rotation $R_α$ in the combined $C^2$ and $B^2_{2,q}$ topology whose normalized conjugacy is not Lipschitz. We also prove that $F''\in B^0_{2,1}$ implies global $C^1$ conjugacy. Thus the thresholds in the Besov summability and weak Dini scales are $q=1$ and $γ=1/2$, respectively. These results complement Herman's Baire category theorem on the residual existence of $C^2$ diffeomorphisms without $C^1$ conjugacy by providing quantitative counterexamples and sufficient conditions for differentiable conjugacy.

math.DS↗

arXiv2026

Beyond Policy Engines: Binding-Complete Execution Admission

Policy engines determine whether evaluated policies permit a request, but a correct ALLOW decision does not establish complete governance-source participation, exact context binding, or sufficient grounds for execution admission. This paper examines the gap between policy evaluation and execution admission. We introduce the Binding-Complete Execution Admission Contract (BCEAC), a governance-relative contract for independently determined source obligations, exact policy and contextual binding, and typed admission outcomes under explicit trust assumptions. We propose a conditionally engine-agnostic architecture in which participating backends must satisfy declared semantic profiles, adapter contracts, and contextual correspondence obligations; an Execution Boundary Language (EBL) authoring syntax remains optional. The architecture separates governance resolution, policy evaluation, independent admission verification, and downstream execution mediation. We investigate two distinct experimental tracks: a finite Governance Source Completeness study and an OPA/Cedar experiment with archived native-engine records examined by offline review. The OPA/Cedar archive comprises 72 scenarios and 144 strict case-engine records (120 native-invoked and 24 pre-native), with eight additional permissive controls outside the strict denominator. The offline review did not rerun the original native OPA/Cedar executables. The results provide bounded evidence while retaining historical failures and unresolved implementation defects. The work does not establish constitutional adequacy, general semantic refinement, authenticated currentness, machine-checked execution-safety theorems, or protected external effects. Its contribution is an explicit, falsifiable framework for determining what must be established between a policy decision and an execution-admission judgment.

cs.CR↗

arXiv2026

Persistence of invariant graphs for twist maps under analytic perturbations

We construct high-frequency perturbations of an integrable exact symplectic twist map which preserve an analytic invariant graph with a prescribed rotation number. Their oscillation frequencies may be any positive integers tending to infinity. For every fixed Gevrey parameter $L>0$, the perturbations tend to zero in a Gevrey-$α$ norm and diverge in a Gevrey-$(α-\varepsilon)$ norm. The frequency class strictly contains the $α$-Brjuno--Rüssmann class, and their difference has cardinality $2^{\aleph_0}$.

math.DS↗

arXiv2026

Knowledge-Guided Construction of Machine Learning Potentials for Reactive Behavior in Energetic Materials

Energetic materials involve multiple interactions, making reliable machine-learning potentials (MLPs) difficult to construct without extensive datasets. Here, we enhance the ability of the neural network interatomic potential (NNAP) to describe molecular systems using sphmulti, a refined spherical-harmonic basis set that combines extra radial basis with element-specific chemical weighting in each radial channel. We further integrate hyperactive learning, farthest-point sampling, and iterative sufficiency criteria with knowledge-guided key events, enabling the NNAP to retain essential information for describing the material system. For 3,3'-diamino-4,4'-azoxyfurazan, the resulting NNAP is trained on only 13,680 configurations yet captures key properties, including the isomerization barrier, pyrolysis activation energy, and shock Hugoniot behavior. NNAP-driven molecular dynamics simulations further reveal that pyrolysis proceeds through furazan-ring opening followed by chain reactions, whereas shock loading induces anisotropic hydrogen transfer driven by interlayer structural deformations. This work enables efficient and reliable MLPs for energetic materials.

cond-mat.mtrl-sci↗

arXiv2026

X-Fly: Learning to Reach and Stop in Full-Chunk Aerial Vision-Language Navigation

Aerial vision-language navigation (VLN) agents built on large vision-language models are expensive to query. Executing a complete predicted action chunk before replanning amortizes this cost, reducing model inference from 195 to 42 ms per executed waypoint for our policy. It also moves the states at which the policy is queried away from expert demonstrations, which supervise neither recovery from these states nor termination at the learner's own arrival. On OpenFly test-seen, no-stop failures account for 47.78% of full-chunk behavior cloning (BC) outcomes. We present X-Fly, a staged training framework for full-chunk aerial VLN. Recovery DAgger labels learner-visited decision states with smooth chunks that reconnect to future expert segments and with STOP at the first in-region state. Flow-GRPO then jointly refines the chunked flow controller and a Bernoulli STOP policy from same-task closed-loop outcomes, supplemented by outcome-verified stopping supervision. With eight controls per decision, X-Fly reaches 74.18% SR on test-seen and 51.73% on test-unseen, compared with 16.45% and 18.43% for full-chunk BC and 32.04% and 29.46% for DreamFly.

cs.RO↗

arXiv2026

Directed impulsive molecular dynamics reveal energy-routing pathways of bond cleavage in TATB

Responses of energetic materials to external stimuli involve more than uniform thermalization. Understanding directional energy redistribution is essential for structural and ignition design. Here, we use TATB as an example to construct a vibrationally faithful machine-learning potential from only 4,630 configurations. To resolve the elusive nonequilibrium energy flow, we develop directed impulsive molecular dynamics, a statistically controlled protocol that repeatedly deposits equal kinetic energy along selected vibrational modes. By amplifying mode-specific responses while tracking the ensuing nonequilibrium evolution, this approach turns energy redistribution into a directly traceable routing process. We find that initial bond cleavage depends strongly on the initially excited mode, rather than being determined by bond-cleavage barriers alone. Specifically, modes that readily trigger reaction efficiently channel energy towards a bond cleavage coordinate, whereas modes that are less effective at inducing break divert energy into soft-like modes or trap it in modes from which energy flow is inefficient.

cond-mat.mtrl-sci↗

arXiv2026

Constraint on particle mass of dark matter based on XTE J1814-338 observation and DANS model

The small observed radius of neutron star XTE J1814-338 (with mass of $1.21_{-0.05}^{+0.05}$ $M_{\odot}$ and radius of $7_{-0.4}^{+0.4}$ $\rm{km}$) poses a great challenge to the traditional neutron star model composed of normal nuclear matter. The dark matter admixed neutron star (DANS) model provides a clue to solve this challenge. By employing the self-interacting fermionic dark matter model, we investigate the effects of dark matter parameters including the particle mass, self-interaction strength, and mass fraction of dark matter on the properties of DANSs. Our results indicate that the joint constraints from the observation of XTE J1814-338 and the astrophysical limits on the dark matter self-interaction cross section derived from observations of different colliding galaxy clusters can provide effective constraints on the particle mass of dark matter. Specifically, to simultaneously satisfy the small radius of XTE J1814-338 and the self-interaction limit, the particle mass of self-interacting fermionic dark matter is constrained to the range from sub-GeV to a few GeV.

astro-ph.HE↗

arXiv2026

On the sharpness of the $C^1$-norm threshold for perturbations in the normally hyperbolic invariant manifold theorem---a toy model perspective

The classical normally hyperbolic invariant manifold theorem asserts that a $C^1$ normally hyperbolic invariant manifold persists under sufficiently small $C^1$ perturbations. For a family of standard-like dissipative twist maps with Jacobian determinant $λ\in(0,1)$, we prove that $C^1$ perturbations of Lipschitz seminorm at most $(1-\sqrtλ)^2$ preserve a unique $C^1$ invariant graph, including at equality. Under the strict bound, this graph is normally hyperbolic. Conversely, for every prescribed $a^*>(1-\sqrtλ)^2$ and every rational frequency, we construct a $C^\infty$ perturbation of $C^1$-norm exactly $a^*$ whose unique invariant graph is Lipschitz but not differentiable everywhere. We also give an analytic perturbation at the critical norm whose invariant graph is $C^1$ but not normally hyperbolic. Thus the closed bound for $C^1$ regularity and the strict bound for normal hyperbolicity are both sharp.

math.DS↗

arXiv2026

Reformulating Pfaffian Quantum Monte Carlo with the Hybrid Monte Carlo formalism

Pfaffian quantum Monte Carlo extends auxiliary field methods to anomalous fermionic Hamiltonians, opening access to pairing systems and Hubbard--Stratonovich decompositions that are unavailable in determinant formulations. We show that this flexibility can be combined with the global updates of Hybrid Monte Carlo (HMC), leading to an algorithm scaling as $\mathcal{O}(N_τL^3)$, similar to its determinant quantum Monte Carlo (DQMC) counterpart without pseudofermions. The central observation is that the molecular dynamics force depends only on the magnitude of the Pfaffian weight, which admits a determinant representation and leads to a simplified force expression. This construction also yields unequal-time Majorana correlators at essentially no additional cost. We demonstrate the method for the interacting Kitaev chain using the imaginary and symmetric Majorana hopping channel. The resulting noncommuting auxiliary-field bond operators are treated through an even-odd decomposition. Their block structure permits an analytic reduction to a sparse effective interaction matrix of bandwidth two and allows explicit force expressions. Results agree with exact diagonalization and resolve low-energy modes localized at opposite boundaries, together with their increasing overlap near the topological transition of the non-interacting system. These findings establish HMC as an efficient sampling framework for Pfaffian auxiliary-field simulations.

cond-mat.str-el↗

arXiv2026

Optimizing Frequent Checkpointing via Low-Cost Differential for Distributed Training Systems

Distributed training of large deep-learning models often leads to failures, so checkpointing is commonly employed for recovery. State-of-the-art studies focus on frequent checkpointing for fast recovery from failures. However, frequent checkpointing generates numerous checkpoints, incurring substantial costs and thus degrading training performance. Recently, differential checkpointing has been proposed to reduce costs, but it is limited to recommendation systems, so its application to general distributed training systems remains unexplored. In this paper, we find that gradients generated during distributed training can be reused to construct differential checkpoints, while the former's size is smaller than the latter's, motivating us to reuse gradients for low-cost differential checkpointing. Based on this main idea, we propose \sysname, a frequent checkpointing framework for compression-enabled training systems that reuses compressed gradients as differential checkpoints, eliminating redundant differential computation and reducing checkpoint transmission cost. Furthermore, we extend gradient reuse to scenarios without gradient compression and propose \sysnameplus, which employs layer-wise-reuse snapshotting and incremental-merging persistence to overlap checkpointing with training execution. Experiments on diverse workloads, including billion-parameter-scale models, demonstrate that \sysname and \sysnameplus significantly reduce checkpointing overhead and enable checkpointing at frequencies as high as once per iteration, reducing training time by up to 89.2\% and 81.2\%, respectively.

cs.DC↗

arXiv2026

Evidence Coverage for Intent-Bound Execution: Scope, Obligations, and Cutoff Reasoning

A verifier may authenticate every available record and still lack grounds to call an execution account complete. Such a claim requires a justified account of which records were due for the execution being assessed. We present an analytical model for retrospective coverage of declared execution-evidence obligations. Its scope binds a structured Intent, an exact Candidate, a selected analytical attempt, an execution and evidence boundary, a stage horizon, a fixed record-obligation profile, a named verifier, and an assessment cutoff. Branch and trigger premises determine obligation instances; source competence, content, integrity, and object and stage bindings determine admissibility. We distinguish closure of the obligation inventory from closure of the relevant verifier view, and define three reporting results: COMPLETE_WITHIN_SCOPE, INCOMPLETE, and UNKNOWN. These results concern current coverage of obligations due at the cutoff. Execution progress, external outcome knowledge, and historical delivery timeliness are reported separately. Constructed service-principal-disablement cases demonstrate complete dispatch and refusal branches, a due but missing final-result record, subsequent coverage after late delivery, and the limits of extending one selected attempt's coverage to all attempts. The contribution is an execution-specific composition of scope, branch, horizon, obligations, admissibility, view, and cutoff. Completeness remains conditional on the declared profile and assessment premises.

cs.CR↗