Papers

Understanding the physical origin of the piecewise-linear computation in deep rectified linear unit (ReLU) networks remains a fundamental challenge. Here we establish an exact statistical-mechanical realization of arbitrary ReLU networks. Starting from a microscopic configuration space and its state multiplicities, we construct a partition function without prescribing a neural-network activation function. The resulting system admits equivalent descriptions in terms of cascaded quantum operations with post selection, or fermionic transport model. We rigorously prove that in $\beta\to+\infty$ limit, the thermodynamic observables of this system exactly reproduce the hidden states, outputs, and loss function of an arbitrary deep ReLU network. Crucially, we demonstrate that the piecewise-linear behavior of the ReLU network emerges from a first-order phase transition in the microscopic system, where the discontinuities precisely coincide with the boundaries of linear regions of ReLU network. This work establishes an exact physical foundation for deep neural computation and provides a novel statistical-mechanical perspective on neural architecture design.

Recent years have seen significant interest in finding equivalent norms on Cesàro function spaces together with their optimal equivalence constants. In this work, we continue this line of research by deriving sharp two-sided inequalities relating the $L^p$-norms of $\mathcal{C}_m f$ and $\mathcal{C}_n f$, where $\mathcal{C}_k$ is the generalized Cesàro operator of order $k \in \mathbb{N}$ acting on non-negative measurable functions $f$ on $(0,\infty)$. A parallel study is also conducted for the corresponding adjoint operators.

This study develops a prescribed-time performance-shaping control method for curvature tracking of a single-segment flexible arm actuated by three antagonistic tendon pairs. A Cartesian curvature representation is introduced to avoid the undefined bending direction at the straight configuration and to establish an explicit six-tendon kinematic mapping. A cubic performance boundary contracts smoothly from an initially admissible error bound to a nonzero terminal accuracy bound within a prescribed time. Based on this boundary, a dual transformation combining static symmetric error scaling and time-varying behavior shaping maps the tracking error into a fixed unit box. The resulting controller guarantees boundary invariance, prescribed-time entry into the terminal accuracy region, and subsequent asymptotic convergence. Numerical evaluations with Python and OpenCR–MuJoCo, together with a supervised reduced-order experiment on a two-section, four-channel platform, provide complementary validation. Across six experimental trials, no violation of the prescribed boundary is observed, and the proposed controller reduces the mean terminal curvature RMSE by 32.5% relative to a matched baseline, with comparable terminal-band entry times. These results support the feasibility of the proposed approach in the reduced-order experimental setting.

A typical first-order ODE syllabus is often presented as a sequence of named methods, which can obscure the unity among them. We propose organizing a substantial part of the classical symbolic first-order syllabus around a single computational target: an exact representative. For a regular scalar equation represented by a one-form $\omega$, the reduction problem is to find, on an appropriate coordinate patch, a coordinate change $\Phi$ and a nonvanishing factor $\mu$ such that $\mu\Phi^*\omega=dG$, so that solution curves are level sets $G=C$. This yields a two-floor architecture. The structural floor records the target, transformations, and domain restrictions; the operational floor executes substitutions, normalizations, and integrations. Linear and separable equations arise from the simplest one-variable integrating factors, while homogeneous, Bernoulli, and related classes become reduction paths to the same target. In an AI- and CAS-assisted environment, this shifts emphasis toward recognition, justification, auditing, and interpretation. The extended version also develops reverse construction, a structural exercise laboratory, comparisons of reduction paths, and a higher-order outlook through first integrals, operator factorization, and structural normalization.

Agentic systems increasingly invoke tools, services, data, and other agents across organizational boundaries, yet a relying party cannot assess a delegated action solely from producing-domain controls and records. This paper develops Trustworthiness as a Service (TaaS) through a synthesis of trustworthy-AI governance, agent security, distributed trust management, identity, provenance, assurance, and control-plane research. The analytical unit is a cross-domain reliance proposition that names the issuer, subject and action, relying party, administrative boundary, evidence dependencies, adverse condition, and required verification or adjudication semantics. The three-condition structural-externalization diagnostic identifies propositions that depend on multiple domains, require producer-independent reliance, and must remain reviewable after revocation, failure, conflicting records, or dispute. For such propositions, the paper specifies a trust-evidence envelope: an immutable workflow manifest linked to append-only, issuer-attributed attestations for task-scoped authority, policy and execution decisions, provenance, validity, disclosure, status, challenge, and recovery. A topology-neutral logical reference model assigns these functions to explicit roles and trust domains. Three analytical scenarios and the TaaS-Eval protocol proposal define manifests, independent consumers, hard gates, adversarial evidence tests, metrics, and reproducible artifact reporting. By composing established identity, authorization, provenance, assurance, and governance mechanisms around a bounded delegated action, TaaS provides a reusable profile for cross-domain reliance. It makes evidence dependencies, independent verification, challenge, and recovery explicit, supporting interoperable governance and future evaluation without treating producer assertions as ground truth.

We consider local cohomology sheaves on compact Hermitian symmetric spaces supported in Schubert varieties, which carry natural structures as mixed Hodge modules. Our main result gives an explicit root-theoretic construction, uniform across Lie types, for the D-simple composition factors and weight filtration on these modules. For Lagrangian and orthogonal Grassmannians, we translate these formulas into combinatorial rules involving strict partitions and shifted Dyck patterns, analogous to existing formulas on ordinary Grassmannians. As an application, we expand upon work of Raicu–Weyman to describe the weight filtration on local cohomology supported in symmetric determinantal and Pfaffian varieties. We also give formulas for the Hodge rational homology level and local cohomological defect of every Schubert variety in a compact Hermitian symmetric space. Finally, we tabulate the local cohomology weight filtrations for all Schubert varieties in the two exceptional Hermitian pairs.

The medications prescribed to a patient often change during a hospital admission as clinicians start, stop, or continue therapies. We study whether models can predict which medication classes are added or removed between 24 hours after admission and discharge. Metrics that compare the complete discharge regimen can reward models for copying medications that remain unchanged, even when they identify no actual changes. We therefore introduce a leakage-controlled benchmark that predicts net ATC3 additions and removals using only prior completed admissions and information available within the first 24 hours of the current admission. Addition candidates are classes not active at 24 hours, whereas removal candidates are classes active at that time. We also introduce R-GEAN, an asymmetric candidate-scoring network with independent addition and removal predictors. Across 240,480 admissions from 82,286 patients, R-GEAN achieves the highest predefined summary of addition, removal, changed-regimen, and action-pattern performance, termed the edit composite (0.464), compared with 0.435 for the strongest primary comparator. Reimplemented RETAIN, GAMENet, and MICRON baselines obtain 0.428, 0.420, and 0.288, respectively. R-GEAN's advantage is concentrated in correctly identifying medication classes no longer active at discharge, while rare additions and admissions with multiple medication changes remain difficult. Rankings based on micro-F1 over the reconstructed discharge regimen and the edit composite correlate weakly across the evaluated models (Spearman r = 0.20). The continuation baseline achieves the highest complete-regimen score despite predicting no additions or removals. These results show that complete-regimen and edit-level evaluation measure different aspects of medication prediction. The benchmark evaluates observed prescribing changes, not treatment appropriateness

In this paper, we study graph editing problems on geometric intersection graphs. For a tuple $\mathcal{S}=(S_1,\dots,S_n)$ of geometric objects in some Euclidean space, let $G_\mathcal{S}$ be their intersection graph. We study the problem of finding a tuple $D=(d_1,\dots,d_n)$ of movement vectors such that the resulting intersection graph $G_{\mathcal{S}+D}$ (after moving, for every $i \in \{1,\dots,n\}$, object $S_i$ by $d_i$) has a predefined property and the total movement distance $|D|$ is minimum. In the weighted version, we are also given a weight vector $w=(w_1,\dots,w_n)$ with positive entries, and the objective is to minimise the total weighted movement distance $|w \cdot D|$. We first consider the property locally dense, which we define as containment of a $k$-clique. Given $n$ weighted intervals, we solve the problem with respect to this property in $O(k^{1/3} n \log^{1+\varepsilon} n)$ time for any $\varepsilon>0$. We then consider $k$-connectivity for $1\le k \le n-1$. Given $n$ unweighted unit intervals, we solve the problem in $O(n^2 \log n)$ time and, for $k=1$, in $O(n\log n)$ time. For $k=1$, we prove strong NP-hardness on intervals of arbitrary length and on weighted unit disks (with only two distinct weights), and weak NP-hardness on weighted intervals (even when lengths equal weights).

This paper investigates the role of quantum grassroots organizations in The Association of South-east Asian Nations (ASEAN), the challenges they face, and how these organizations can be better supported by other players in the ecosystem of quantum technologies. We look into the goals and strategies of ASEAN grassroots organizations and their potential impacts and investigate the obstacles and challenges currently faced by these groups. We propose providing mentors with specialized knowledge to assist in the refinement of technical content and organizational management, building connections between grassroots and key stakeholders in the quantum space, and creating opportunities to obtain stable funding. These can help to mitigate the aforementioned challenges and ensure its longevity, sustainability, and impact on the wider quantum community.

Parkinson's disease alters gait and bilateral coordination, but machine-learning performance also depends on how continuous gait signals are represented. This study investigates whether preserving anterior-posterior center-of-pressure (AP-COP) information at fixed locations across normalized stance provides a compact and informative representation of plantar-force gait signals. Bilateral vertical ground reaction force recordings from 165 participants in the Gait in Parkinson's Disease Database were evaluated using repeated fully nested participant-level cross-validation. We propose AP-COP10, comprising AP-COP position and bilateral asymmetry across five stance windows. AP-COP10 achieved an AUC of 0.894 and outperformed three harmonized literature-derived COP representations under the same evaluation pipeline. The complementary 25 non-AP-COP descriptors alone achieved an AUC of 0.856, while the complete 35-feature representation achieved 0.908. Removing AP-COP10 from the complete representation produced a statistically supported loss in discrimination, whereas adding the complementary descriptors to AP-COP10 yielded only a small, unsupported improvement. Feature competition indicated that the most informative stance-indexed descriptors were concentrated in early and early-mid stance, while source-study holdout and sensor-perturbation analyses supported the robustness of the representation. These findings indicate that stance-indexed AP-COP retains discriminative information that is not readily recovered by broader engineered gait descriptors, supporting compact and interpretable representations for machine-learning analysis of pathological gait.

We extend the high-order Zipped Finite Element Method to three-dimensional polyhedral elements. The construction relies on a local sub-tetrahedralization obtained by triangulating the polygonal faces and connecting the resulting vertices to a suitable interior point. High-order shape functions are constructed as linear combinations of standard finite element basis functions, with coefficients chosen to ensure exact polynomial reproduction on the faces and in the element interior while preserving global C0-conformity. We further introduce a QR-based strategy as a systematic alternative to the heuristic procedure that was adopted for the two-dimensional case. The QR-strategy is more systematic and robust than the heuristic strategy, while being independent of the spatial dimension and avoiding geometric alignment tests and problem-dependent tolerances. The resulting method is formulated for a diffusion-reaction problem and numerically assessed through three-dimensional convergence tests, confirming the expected optimal polynomial order of accuracy.

Mobile gas-sensing inspection requires a mobile platform to reliably reach ordered sampling poses. We present MIRA-PRM, a mission-informed probabilistic roadmap that combines geometry-, mission-flow-, and inspection-conditioned sampling with locally adaptive connectivity and evidence-gated refinement. Eight development campaigns comprising 142,448 planner runs evaluated mission scaling, repeated-query adaptation, finite node budgets, target distributions, local map repair, ablation, cross-family comparison, and parameter sensitivity. MIRA-PRM maintained 100% seed-level success as ordered targets increased from two to eight, and achieved 97.67% seed-level success under a 40-node ceiling, compared with 67.17% for PRM and 86.50% for PRM*. Local repair reduced event time by 79.9-81.9% and node additions by 90.8-94.3% relative to rebuilding. On a separate frozen holdout of 12 maps, 48 ordered missions, and three seeds per map-mission unit, MIRA-PRM produced 96/144 validated successes, versus 82/144 for PRM, 89/144 for PRM*, and 70/144 for LE-HG-PRM. On paired common-success runs, MIRA-PRM reduced wall time by 16.89% relative to PRM* and 81.01% relative to LE-HG-PRM, but was 52.23% slower than PRM; paths were 0.54-1.86% longer. All validated paths passed independent full-segment checks and clearance non-inferiority. The results support a reliability-oriented trade-off in structured simulations and motivate external-map and physical mobile-sensing validation.

We study ground states of the following Dirac-Coulomb system $$\mathscr{D}_cu_c- \left(\abs{x}^{-1}\ast \abs{u_c}^2\right)u_c=\omega_cu_c,$$ under the constraint $\norm[L^2]{u_c}^2=1.$For sufficiently large \(c\), we prove uniqueness of ground states modulo translations, phase transformations and spinorial rotations, and determine the kernel of the linearized operator at ground state. We also obtain asymptotic expansions of ground state in the nonrelativistic limit regime.

We derive representation-free densities of ratios of the positive stable random variable and obtain their Laplace/Stieltjes transforms. We state representation-free densities of occupation times (Lamperti, 1958) and Bessel process excursions (Bertoin et al., 2006) and their transforms. We also obtain simple representations of these and other densities using a known infinite series representation of the stable density.

Enterprise agentic systems that route every trajectory step to a frontier model waste 60-80% of their inference budget on subtasks that smaller models handle equally well. Existing routing solutions optimize single-turn query assignment but ignore a property unique to agentic workflows: subtask complexity varies widely within a single trajectory. A planning step may require frontier-class reasoning while a subsequent formatting step needs only a 7B model. We formalize step-level model routing as a sequential assignment problem over agent trajectories and propose AgentRouter, a lightweight classifier (12M parameters, <5ms overhead per step on an A100 GPU) that maps each trajectory step to one of four model tiers using five features extractable at routing time. Trained on 50,000 annotated agent trajectory steps spanning planning, coding, research, and data analysis tasks, AgentRouter achieves 72% cost reduction relative to frontier-only baselines, retaining 97.3% of frontier-only quality (less than 3% degradation in end-to-end task completion); per-step routing accuracy reaches 91% on minimal-complexity steps and 85% on efficient-tier steps, with 76-82% on the harder mid-range and frontier tiers. On the same benchmarks, RouteLLM and FrugalGPT (applied per-step) achieve only 31% and 44% cost reduction respectively, because their single-turn training signal misses trajectory-level quality dependencies.

While compressed-sensing regularizers enable interpretable reconstruction of 4D Flow CMR through transparent variational objectives, their hand-crafted nature is too restrictive under high acceleration. State-of-the-art learning-based approaches mitigate this, but typically encode regularization implicitly through unrolled network modules, which limits their interpretability. To address this limitation, we propose CLEAR, designed to combine the interpretability of compressed sensing with the flexibility of learned models. To the best of our knowledge, it is the first learned regularizer for a 4D reconstruction task. In the ultra-accelerated \(10\times\)–\(50\times\) regime of the CMRx4DFlow2026 challenge, CLEAR outperforms compressed sensing locally low-rank (LLR) and the popular variational network FlowVN, while using less than 10k parameters and preserving an interpretable regularization structure.

Existing prompt injection research focuses on single-model chatbot scenarios, where an attacker manipulates one LLM through crafted input. Multi-agent systems amplify this threat through three mechanisms absent from single-model settings: inter-agent message passing creates injection channels invisible to perimeter defenses, shared tool access enables privilege escalation across agent boundaries, and trust propagation allows a compromised agent to influence upstream orchestrators. We construct a threat model enumerating 14 attack vectors across four categories: direct injection via user input (3 vectors), indirect injection via tool outputs (4 vectors), inter-agent injection via message passing (4 vectors), and cascading injection through orchestrator manipulation (3 vectors). Testing all 14 vectors against a 6-agent production-representative system, we find that 67% of agents are vulnerable to at least one scope violation even with system-prompt-level guardrails, and indirect injection via tool outputs succeeds in 43% of attempts. Four architectural defenses reduce overall injection success from 31.2% to 4.2%: message signing with provenance tracking (inter-agent injection down 91%), input/output sanitization at agent boundaries (indirect injection down 78%), privilege-scoped tool access per agent role (privilege escalation eliminated entirely), and anomaly detection on inter-agent communication patterns (84% of cascading attempts caught).

We present the first study of shear viscosity induced cavitation incorporating transverse expansion alongside chemical non-equilibrium dynamics in the Quark-Gluon Plasma (QGP). Chemical non-equilibrium is incorporated through quark and gluon fugacities, whose evolution is coupled to causal dissipative relativistic hydrodynamics in the Gubser geometry. Using phenomenological temperature-dependent shear viscosity parametrizations, we study the evolution of the longitudinal pressure and identify the conditions under which it becomes negative and thereby breaking down the hydrodynamic description. We demonstrate that shear induced cavitation develops first at the centre of the fireball ($r=0$) at early times and subsequently extends towards larger radial distances, with transverse expansion accelerating its onset, while persisting during chemical equilibration. Within our framework, we determine the critical initial shear viscosity $(\eta/s)_{\rm crit}$, below which cavitation does not occur throughout the QGP evolution. Since hydrodynamics is found to successfully describe the phenomenology of heavy-ion collisions, our results provide a constraint on shear viscosity requiring the absence of cavitation during the QGP evolution. We find that the phenomenologically extracted $\eta/s$ from JETSCAPE analyses lie below the critical shear viscosity $(\eta/s)_{\rm crit}$ obtained in our framework.

Reinforcement learning (RL) is vital for optimizing video generation models, with a robust reward model (RM) serving as the cornerstone. However, existing video reward models often produce unstable scalar scores because they directly map complex, subjective video quality into a single score without explicit evaluation criteria. This leads to scalar drift, where the scoring scale collapses or shifts across different prompts, making the reward unreliable for RL. Drawing inspiration from professional human annotation engineering, we address this problem with RewardVerse, a rubric-based video reward framework that introduces a dynamic rubric as an intermediate representation between the evaluation query and the scorer. Instead of unconstrained direct scoring, RewardVerse first generates explicit evaluation criteria and then performs rubric-guided scoring, providing a stable semantic anchor that mitigates scalar drift. To efficiently optimize this collaborative pipeline, we propose Rubric-Guided Policy Optimization (RGPO), a two-stage training algorithm. RGPO first warms up the scorer using self-evolving seed rubrics and then jointly optimizes the rubric generator to produce query-adaptive evaluation criteria while continuously aligning the scorer with human ratings. Extensive experiments on the 16-dimensional EvalVerse benchmark and external datasets demonstrate that RewardVerse mitigates scalar drift, achieves state-of-the-art performance on both pointwise and pairwise evaluation, and provides a robust and interpretable reward signal for RL in video generation.

Quantum key distribution (QKD) can deliver information-theoretic security, but its receivers are still typically assembled from cascaded bulk optics. Metasurfaces offer a radical reduction in this complexity. Here, we make the metasurface function as the QKD measurement device itself. We design wavelength-specific dielectric apertures at 780, 1550, 2000, and 10 600 nm that interweave the linear H/V and circular R/L phase libraries at a 0.44–0.49{\lambda} cell pitch, i.e., below half the operating wavelength, and map the complete passive-BB84 measurement directly onto four detector channels. This subwavelength basis co-location prevents an ordinary propagating-wave aperture mask from isolating one basis without simultaneously attenuating or perturbing the other. Simulated full-wave focal-plane intensities are converted directly into conditional Born probabilities using fitted, calibration-fixed detector regions. The selected designs achieve probability fidelities of 97.96–99.08%, four-port collected efficiencies of 17.43–40.94%, and mean intrinsic quantum bit error rates of 1.42–3.96%. At 1550 nm, the selected design gives an asymptotic device-level yield of 0.151 secret bits per incident photon under an ideal single-photon model. Six-state detector tomography gives positive both-basis security bounds for all six reconstructed receiver models, including their polarization-dependent loss, basis imbalance and crosstalk. These results establish protocol-matched meta-optics as a route to compact, passive QKD receivers.

We present the validation and dynamical characterization of TOI-789 (TIC 300710077), a compact multi-planet system around an M3 dwarf at 43 pc. Using ~7 yr of TESS photometry (42 sectors), public ground-based follow-up (LCOGT, MEarth), and high-resolution imaging (Gemini/Zorro, SOAR), we statistically validate three terrestrial planets with TRICERATOPS: TOI-789 b (Rp = 1.18 Rearth, P = 5.45 d), c (1.15 Rearth, 8.04 d), and d (1.40 Rearth, 12.97 d), with false-positive and nearby-false-positive probabilities below the validation thresholds once resolved neighbours are excluded using the on-target ground-based photometry. Through an iterative Transit Least Squares search we identify a fourth, previously uncatalogued planet candidate (P = 19.74 d, Rp ~ 0.88 Rearth), the outermost and most temperate member (S = 1.71 Searth, Teq ~ 292 K); its low signal-to-noise ratio precludes formal validation, so we report it as a candidate pending confirmation. A dynamical analysis (SPOCK and N-body integrations) indicates a dynamically cold, near-resonant system, stable only for low eccentricities (e < 0.05), eccentricity rather than mass limiting stability; including the fourth candidate tightens this further. The four objects span the Earth-Venus transition across the Venus zone, two validated planets falling in the temperate regime under the equilibrium-temperature criterion, making the system a potential comparative-planetology laboratory anchored on confirmed planets. The low TSM (~3-5) and moderate host brightness place atmospheric characterization beyond reach. TOI-789 exemplifies the numerically dominant population of compact terrestrial-planet systems around M dwarfs whose characterization is limited by host brightness, for which dynamical characterization offers a complementary route.

Autonomous AI agents increasingly act across organizational boundaries on behalf of human operators: they invoke third-party services, delegate subtasks to other agents, and pay for metered resources. Deploying such agents safely requires five capabilities that today live in separate systems: persistent identity, scoped delegation, peer trust, discovery, and payment. Existing approaches root these in centralized authorities or cover only subsets, so authority, trust, and payment fracture exactly where autonomy needs continuity: when a key rotates or a delegation must be revoked. We present NostrAgent, a decentralized architecture that unifies all five over Nostr relays using three custom event kinds: Kind 38100 identity declarations authenticated by BIP340 Schnorr signatures with pre-rotation commitments, Kind 38101 scoped delegation chains whose every hop verifiably narrows granted capabilities, and Kind 38102 peer attestations forming a Sybil-deterrent trust graph, with Lightning HTTP 402 (L402) binding payment to agent identity. Identity remains operator-sovereign without any registration authority; relays are substitutable transport rather than a trust root; and every authorization decision is replayable offline from signed events. We evaluate a Python prototype with a mixed-method design: ATAM quality analysis with a two-round mini-Delphi panel, STRIDE threat modeling across three trust boundaries, eleven benchmarks with non-parametric statistics, and 19 failure modes. Results show sub-millisecond offline verification, linear delegation-chain scaling, and Lightning-settled L402 at 157 ms median on regtest. 17 of 19 failure modes pass empirically, one is bounded analytically, and one is disclosed as an architectural limitation. NostrAgent demonstrates an auditable prototype substrate for trustworthy agentic systems without centralized trust roots.

Token selection for instruction tuning is coupled: supervision utility depends on retained context, and vice versa. Existing methods largely optimize prompt compression or response supervision separately. Combining selectors can value supervision using removed context, or context using excluded supervision, creating selection-state mismatch. BRIDGE (Budgeted Response-Prompt Interaction via Directional Gradient-guided Efficient Token Selection) models selection-conditioned token utility through a shared validation-directed surrogate objective. BRIDGE routes response utility to supporting prompt tokens through attention, then alternates budgeted selection to revalue each side under the other's retained subset. Updates reuse precomputed interactions without repeated model scoring, while structure-aware projection promotes coherent supervision spans. Across three model families and three task domains, BRIDGE performs best overall among evaluated selection methods. Its MATH advantage over independent selection grows with compression, reaching 1.66 percentage points on Llama-3.2-3B.

Generative models are rapidly expanding image quality assessment (IQA) beyond traditional fidelity factors to emerging dimensions such as physical plausibility and text-rendering correctness. However, existing IQA models rely on fixed definitions and heavy supervision, making them difficult to extend to open-ended perceptual dimensions. We identify holistic bias as an important limitation: when scoring an unseen dimension, models reuse generic quality priors, leading to scoring errors and rank inversion. To address this, we propose PACE (Perceptual Agentic Collaborative Evolution), a training-free multi-agent framework that formulates open-ended IQA as explicit protocol construction. Given a target dimension, PACE uses collaborative agents to construct an evaluation protocol composed of verifiable Visual Question Answering (VQA) probes, grounding evaluation in concrete visual evidence rather than holistic impressions. The resulting protocol is calibrated using only four human-annotated images per dimension, while a dual-track scoring mechanism aligns model perception with human scoring scales. Across traditional IQA, structural fidelity, context-aware aesthetics, and newly defined open-ended dimensions, PACE consistently improves its MLLM backbone, achieving competitive performance across diverse IQA settings, and reduces the Holistic Override Rate (HOR) from 44.4% to 8.6%.

Novel view synthesis from sparse inputs remains challenging for 3D Gaussian Splatting (3DGS) due to ambiguous geometry, cross-view inconsistency, and missing details in under-constrained regions, resulting in degraded reconstruction and unstable rendering. To tackle these issues, we propose D$^{3}$GS, a Depth-DINO-Diffusion guided sparse-view Gaussian reconstruction framework that jointly enhances geometry and appearance. D$^{3}$GS first recovers a high-resolution, metric depth map via diffusion-based completion and DPT (Dense Prediction Transformer) refinement, providing robust Gaussian initialization and geometric constraints. Then, a DINO-guided view-consistent learning is introduced to augment Gaussian attributes with structural features, improving multi-view consistency. Finally, a diffusion-based Gaussian refinement module injects generative priors into an iterative optimization strategy, enhancing high-frequency geometric and appearance details within the Gaussian representation. Experiments on DTU, LLFF, and Mip-NeRF 360 show that D$^{3}$GS achieves consistent and substantial improvements over strong baselines, with ablation studies validating the effectiveness and complementary roles of each component.

Given a directed graph with positive edge weights and two vertices s,t, a next-to-shortest s-t path is a shortest simple s-t path among those whose length is strictly larger than the shortest-path distance. The problem was introduced by Lalgudi, Papaefthymiou and Potkonjak in 1996; it is NP-hard when zero-weight edges are allowed, and its complexity on positively weighted digraphs remained open for almost three decades until Chen, Wein and Zhang recently gave a polynomial-time algorithm running in O(n^4 m^3 log n) time. We give a substantially faster algorithm within their optimal-middle-segment framework. The core idea is to split the problem into "choosing a prefix" and "completing it". Given a prefix P: s -> A made of shortest-path edges, delete the vertices used by P, forbid leaving A along shortest-path edges, and the best completion is one shortest-path computation. The difficulty lies in choosing P: even for a fixed A, deciding whether some shortest prefix admits a completion is NP-complete. We do not solve these fixed-A subproblems one by one. Fix any globally optimal next-to-shortest path; its middle segment induces a boundary edge x -> c in the shortest-path DAG. For the correct triple (A,B,x), the optimal path certifies c as a feasible next hop, and we prove that every feasible next hop that is not earlier than c in a topological order can be combined with the same middle segment into another globally optimal path. Hence only the feasible next hop of maximum topological index is kept per triple, giving O(n^3) representatives, all generated by a two-dimensional DAG dynamic program with a local reward. The total running time is O(n^3 (m + n log n)), and O(n^3 m) on unweighted graphs. The proof rests on an uncrossing lemma: the last intersection between a reference prefix and the candidate's partner suffix can always be moved strictly earlier, which cannot go on forever.

Long-context models read a novel the way a person reads a printout: one token after another, in narrative order, with the whole history competing for a fixed budget of attention. A detective does not work that way. They sort what happened when, and they keep a map of who relates to whom, so a clue from chapter one can meet a question asked at the end of the book. We test whether a frozen knowledge graph can give a small local model that same freedom. Thirty detective novels and 234 multiple-choice questions are answered by one fixed qwen3.5:9b reader under nine conditions: five graph routes, a recent-window baseline, whole-book compression, ordinary vector retrieval, and a question-only control. The strongest graph route reaches 53.85% (126/234) against 46.15% for the recent window, 51.28% for compression, 51.71% for vector retrieval and 40.17% for question-only. On the subset that no model can answer without the book, the graph route reaches 42.86%. None of the fifteen graph-baseline contrasts survives Holm correction, so we present the result as exploratory evidence about a design. Two structural findings survive scrutiny better than the headline number: annotated evidence concentrates in the topological core of these graphs (2.35x enrichment, pooled), and the two graph-building pipelines differ so much in annotation coverage (16% versus 73% of clue paragraphs) that pooled accuracy alone would hide which bottleneck is being measured.

We present reference implementations of four published primal heuristics for mixed-integer programming inside the open-source solver HiGHS (Huangfu and Hall, 2018): Feasibility Jump (Luteberget and Sartor, 2023), fix-propagate-repair (FPR; Salvagnin, Roberti and Fischetti, 2025) with its LP-guided dive-time variant, LocalMIP (Lin, Zou and Cai, 2024), and the first-order-LP feasibility pump Scylla (Mexi et al., 2025). They share one integration interface with per-heuristic budgets and patiences, and each runs in parallel on independently seeded workers. The evaluation is on the mipfeas benchmark (Bussieck and Dirkse, 2026), 233 instances at 600 s, against a separately built unpatched HiGHS on the same machine. The primal integral improves by 23.7% over the full set and by 12.9% on the instances not used for tuning: the patched solver reaches a good solution sooner. At the time limit it is slightly ahead. It finds a solution on 4 instances where the unpatched solver finds none, against 1 the other way, and its final gap is smaller, significantly so only over the full set. Scylla and the LP-guided dive produced no accepted incumbent in any full-limit run that carried them and ship disabled; their implementations remain. The implementation is released as open source.

Negatively charged nitrogen-vacancy (NV$^-$) centers located a few nanometers below the diamond surface are key quantum defects for nanoscale sensing of external spins. However, the creation of shallow NV$^-$ centers with high yield remains a materials challenge. Here, we demonstrate that high-angle ion implantation enhances the creation efficiency of shallow NV$^-$ centers. By implanting $^{15}$N ions at angles exceeding 60$^\circ$, we achieve high NV$^-$ yields approaching 10% with effective NV$^-$ depths below 10 nm. These yields are significantly higher than those typically reported for shallow NV$^-$ creation. The enhanced NV$^-$ yield is consistent with an increased vacancy-to-nitrogen ratio in the near-surface region, which is expected to promote NV formation during annealing. The created NV$^-$ ensembles show coherence properties comparable to those of single shallow NV$^-$ centers at similar depths, and allow detection of nuclear spins in van der Waals materials attached to the diamond surface. Our results establish geometric control of ion implantation as a simple and broadly applicable approach to engineer shallow vacancy-related quantum defects.

Prior studies on wideband physical layer security (PLS) have mostly focused on either far-field or near-field communication systems. In this paper, we investigate PLS in a more general and practical mixed near-field and far-field wideband communication scenario, where a base station (BS) equipped with an extremely large-scale array (XL-array) serves multiple legitimate users in the far field, while a near-field eavesdropper attempts to intercept confidential information at close range. Specifically, we formulate an optimization problem to maximize the secrecy rate of all legitimate users across all subcarriers by jointly designing the transmit beamforming and artificial noise (AN), subject to per-subcarrier power constraints. To gain useful insights into the unique characteristics of mixed-field wideband PLS, we first analyze a special case with a single legitimate user and an eavesdropper. Interestingly, we show that the secure transmission regions in mixed-field wideband systems are not the same across all subcarriers, as revealed by subcarrier-specific secrecy conditions derived from the Fresnel integrals. Moreover, we theoretically demonstrate that introducing frequency-selective AN into the system offers two key benefits: 1) it transforms originally insecure subcarriers into secure ones, and 2) it substantially improves the overall secrecy performance. For the general multiuser setup, we propose an efficient two-stage hybrid beamforming algorithm to jointly design the transmit beamforming and AN, ensuring secure transmission under mixed-field conditions. Building on our analytical insights, we further develop an efficient and low-complexity algorithm that designs the AN beamformers by directly aligning them with the eavesdropper's channel, thereby eliminating the need to optimize AN beamformers across all subcarriers.