Papers

Multi-modal image fusion (MMIF) aims to form a single image by integrating shared information, preserving complementary cues, and coordinating cross-modal conflicts across modalities. However, due to the absence of ground-truth fused images, existing MMIF supervision commonly uses spatial-domain sources or gradient variants as surrogate ground truths, making the supervision mechanism inherently misaligned with the goal of MMIF and causing pixel-level compromise or modality bias. To address this, we propose a relation-constrained supervision paradigm that moves fusion supervision from the spatial domain to a learned relation space. Instead of forcing the fused image to approximate the sources, we use frozen pretrained representation models as information providers and design a learnable feature adapter to align heterogeneous DINO and CLIP features into a unified supervision space. The adapter infers three relation parameters, namely sharedness, dominance, and coordination radius, which define three losses corresponding to the MMIF's goal. To make this space reliable, we devise a self-supervised contrastive ranking objective tailored to the adapter and couple it with the fusion network through alternating optimization. Extensive experiments show that the proposed supervision space yields significant gains regardless of which mainstream backbone the fusion network adopts, offering a supervision paradigm better aligned with the goal of MMIF. Code: github.com/GMY628/RCS-Fusion.

Large online programmes receive heavy volumes of queries during onboarding, at a scale that grows faster than the number of staff available to answer them. This paper reports a study of an AI integrated query resolution platform that spreads incoming queries across four routes: an AI based assistant, answers from fellow participants, a curated corpus of frequently asked questions, and escalation to administrators. Over nine weeks, from 2 May to 6 July 2026, the system handled 4,093 queries raised by 1,434 participants. Nearly every query reached a recorded resolution, and one query in five closed within an hour. Reuse of 114 corpus entries absorbed 21.3% of the volume, participants resolved a further 12.2% on their own, and 132 participants answered questions for one another at a median of 9 to 15 minutes, showing that peer answering, where it occurred, was fast and broadly shared across the cohort. Classifying the query text shows that demand was narrow rather than varied. A single process step, the submission of a certificate and the offer letter that follows it, accounts for 56.8% of corpus mediated resolutions, and at least 20.4% of queries concern the progress of a pending submission rather than a request for information, a class the assistant served only 1.3% of the time, since a stored answer cannot report an individual's current status. Only 8.0% of the queries handled by a person duplicated content already in the corpus, which indicates that the knowledge base was already well used. Most direct administrative closures occur in synchronous bulk events, a pattern that shapes how the records should be read.

We present Video2SwimFish, an automated pipeline and benchmark for building controllable fish assets from real-fish videos for underwater embodied AI. Given synchronized multi-view videos of an individual fish, the pipeline reconstructs a metrically scaled deformable mesh from a VLM-selected canonical frame, generates internal articulation adapted to that individual's morphology through a VLM actor-critic loop, and extracts a Biological Locomotion Manifold (BLM) from the fish's observed midline curvature. The BLM provides a low-dimensional action space bounded by real-fish motion, enabling an individual swimming policy to be learned for each reconstructed fish. We release two paired datasets: synchronized top- and front-view recordings of 120 individual fish across 6 species, and the controllable assets and individual swimming policies derived from them. Because every asset is tied to the animal it came from, the dataset supports a benchmark that evaluates locomotion learning not only on task success but on fidelity to that individual in trajectory shape, body curvature, and tail-beat frequency, across trajectory following, reward-free swimming behavior transfer from video, and a downstream case study in which a simulated BlueROV underwater robot captures one of the assets. We find that task success and locomotion fidelity do not necessarily improve together: the method achieving the highest task completion is not the method achieving the highest locomotion fidelity, and we identify faithful reproduction of individual animal locomotion as an open challenge for the community. Project website: https://hangongchen.github.io/video2swimfish-web/.

We develop a positive-conservative Fourier optimization (PCFO) method for computing ground states of Bose–Einstein condensates with higher-order interactions. The ground-state problem admits a convex density formulation, but the singular behavior near zero density poses difficulties for high-accuracy computation. We introduce convex regularizations of the density energy and establish their $\Gamma$-convergence on the computational domain as the regularization parameters vanish. The regularized problem is discretized by a Fourier pseudospectral method with mass conservation and nodal positivity. For the resulting constrained optimization problem, we use an accelerated first-order method for the main energy reduction, followed by a Newton refinement to reduce the remaining optimality residual. Numerical experiments show spectral-type convergence for regularized problems, quantify the effects of the regularization parameters on accuracy and spatial resolution, and demonstrate the effectiveness of the two-stage optimization method.

Multi-agent debate (MAD) is often used to improve large language model (LLM) reasoning, but sequential debate is rarely a neutral aggregator of agents' opinions. We show that sequential MAD suffers from a pronounced first-speaker bias: agents disproportionately shape the final answer when they speak first. As a result, placing a stronger model after weaker ones can substantially offset its reasoning advantage. We then focus on the disadvantaged strong-agent-last setting and ask whether personality prompting can mitigate this imbalance. Drawing on the Big Five model, we study agreeableness and extraversion as behavioral interventions applied to either the strong or weak side. We find that their effects are trait-specific. Influence consistently shifts in the direction of lower agreeableness, and assigning low agreeableness to the stronger agent helps restore its lost influence and improves final accuracy. Extraversion, by contrast, produces less systematic changes in influence and accuracy, with its clearest effect appearing in agents' verbosity. These findings show that effective MAD design depends not only on model capability, but also on how speaking order and induced interaction behavior shape the debate process.

We present an extension of the Sartre event generator that implements inclusive diffractive deep inelastic scattering for both proton and nuclear targets at small $x$. The calculation is performed in the colour-dipole picture using the IPsat and IPnonsat dipole amplitudes that Sartre also employs for exclusive processes, and includes the transverse and longitudinal $q\bar q$ contributions together with the $q\bar qg$ Fock state that dominates at large $M_X$. The partonic diffractive system is hadronised with \pythia, giving fully exclusive hadronic final states. We compare to the combined H1 and ZEUS inclusive diffractive measurements and to diffractive $D^*$ production, and present predictions for the Electron-Ion Collider. This makes Sartre the only event generator providing inclusive diffractive DIS in small-$x$ for both $ep$ and $eA$ collisions.

Recent work on hallucination detection in large language models has shown that, for a fixed pre-trained model and reasoning task, it is possible to estimate the model's confidence in the correctness of its outputs. Such uncertainty estimates have primarily been used to improve truthfulness by detecting or filtering confabulations. In this work, we ask whether these signals can instead be used more proactively to directly improve the accuracy of model-generated answers. We propose USteer, a simple, training-free steering mechanism that adjusts a model's layer-wise activations during inference using the gradient of a confidence measure with respect to the activations. This procedure nudges generation toward outputs with lower uncertainty at inference time, without modifying model parameters or requiring additional supervision. We show that this approach consistently reduces hallucination across a range of tasks, demonstrating that confidence signals can be leveraged not only for detection, but also for effective inference-time control of model behavior.

We introduce action integrals for partially integrable Hamiltonian systems that interpolate between the non-integrable and completely integrable theory. We show that under the assumption of ergodicity of the Hamiltonian system on the common level sets of its integrals, these actions become constants of motion for the averaged dynamics and, consequently, adiabatic invariants when the Hamiltonian function is allowed to slowly change with time.

This paper presents a scalable optimal-transport-based framework for terminal distribution matching in multi-agent systems. While optimal transport provides a natural way to measure distributional mismatch and assign agents to a desired spatial distribution, global discrete transport can become computationally expensive for large-scale systems. We address this bottleneck by partitioning agents and target samples into spatially corresponding blocks and solving smaller local transport problems. Under a mass-balance condition, the resulting restricted coupling remains feasible for the global problem and provides an upper bound on the Wasserstein cost. The local assignments generate target locations for finite-horizon agent control, applicable to both linear and nonlinear dynamics. By alternating local assignment and control, we establish a cycle-to-cycle descent guarantee for the resulting transport surrogate. The proposed framework therefore enables scalable terminal distribution matching while retaining a rigorous connection to the Wasserstein objective. The technical soundness of the proposed results is validated through simulations.

Real-time classroom polls are widely used and increasingly generated with automated assistance, yet the questions themselves are rarely evaluated as measurements. We audit a large corpus of authentic classroom polls, 604 items across 47 sessions answered 340,668 times by 2,807 learners, as a measurement instrument. For the 539 items whose correct answer could be established and verified from the lecture transcript, we place every item and every student on a common scale using item response theory and analyse the answer structure of the True/False items. Two findings emerge. First, the polls form a coherent but easy scale of moderate precision (marginal reliability about 0.60), on which roughly a quarter of items barely separate stronger from weaker students. Second, students show a robust tendency to answer True, present at the individual level (77% of students lean True), which meets a milder tendency for items to be keyed False; as a result answer direction predicts difficulty, False-keyed items being about thirteen points harder, and the effect survives controls for item content and for selective answering. Both findings rest on signals a polling system already records, so the same checks can be run as items are generated, before they reach students.

Large language models (LLMs) have been rapidly improving in long-context tasks, powered by Chain-of-Thought (CoT) reasoning. However, the internal mechanisms underlying this improvement remain unclear. We investigate these mechanisms through a needle-in-a-haystack (NIAH) counting task, where an LLM is asked to count the number of records dispersed in a long text. Across twelve model comparison groups, Thinking (or reasoning) improves counting accuracy over Non-thinking, with pronounced gains at larger counts. This motivates our mechanistic analysis, which identifies two contrasting mechanisms: (i) broad retrieval, where Non-thinking models broadly attend to multiple needles; (ii) targeted retrieval, where Thinking models use enumeration in CoT traces to successively retrieve needles. Targeted retrieval concentrates attention on individual needles and is accompanied by more compact internal representations. Moreover, causal intervention analysis suggests that Thinking models use the CoT trace to maintain and update an internal counter as needles are successively retrieved, even without explicit numbering. In small controlled experiments, both retrieval mechanisms and counter states emerge under standard autoregressive training. Together, our results connect long-context retrieval with representation geometry of counting, supporting a state-tracking account of CoT reasoning.

The commuting local-Hamiltonian (CLH) problem is a restriction of the local-Hamiltonian problem, in which the terms of the Hamiltonian are required to pairwise commute. A long line of work has shown that the problem lies in $\mathsf{NP}$ for certain families of commuting local Hamiltonians. Nevertheless, there has been no formal evidence against the possibility that the general CLH problem is $\mathsf{QMA}$-complete. CLH is complete for the complexity class $\mathsf{QIMA}$, defined through quantum verifiers whose local gates commute. Therefore, CLH is $\mathsf{QMA}$-complete if and only if $\mathsf{QIMA}=\mathsf{QMA}$. In this work, we introduce a classical-oracle analogue $\mathsf{QIMA}^{\mathcal{O}}$ and construct a classical oracle $\mathcal{O}$ such that $\mathsf{BQP}^{\mathcal{O}}\not\subseteq\mathsf{QIMA}^{\mathcal{O}}$. Since $\mathsf{BQP}^{\mathcal{O}}\subseteq\mathsf{QMA}^{\mathcal{O}}$ for any classical oracle $\mathcal{O}$, this implies $\mathsf{QIMA}^{\mathcal{O}}\neq\mathsf{QMA}^{\mathcal{O}}$ for our constructed oracle. Thus, our result provides relativized evidence against the possibility that the general commuting local-Hamiltonian problem is $\mathsf{BQP}$-hard, and hence also against the possibility that it is $\mathsf{QMA}$-complete.

Routing signals of modern vision transformers – expert gates, attention-residual weights and halting scores – often improve probes that predict whether the model is correct, and the improvement is commonly read as evidence that routing carries information about errors beyond the model's outputs. We test this inference directly: keeping real output-routing pairs, we redraw correctness labels from a frozen output-only generator fitted on disjoint data, so that routing is uninformative by construction. Under this exact label null, a width-matched MLP comparison still reports a routing gain in 51.3% of confidence-only evaluations (308/600), while a linear comparison reports none. Holding each training trajectory fixed on a six-model panel and selecting the checkpoint by validation log loss instead of validation accuracy removes the detections (50/120 to 0/120, and 83/120 to 0/120 in an independently implemented probe), identifying accuracy-based checkpoint selection as the cause; across all output views the raw detection rate falls from 27.5% (528/1,920) to zero observed detections. The repaired comparison is not sensitive, detecting an implanted signal of about 0.005 nats in 0/20 replicates in each of two matched settings, whereas a conditional permutation test built on an estimated routing law detects it in 11/20 and 10/20 and rejects rarely under the null. On real correctness labels, the conditional analysis yields model-relative evidence in five DeiT attention-residual families; in four it persists under two specified variants of the conditional law, and no family passes an additional noise criterion. Fitting a better probe and testing for incremental information are different problems, and each needs its own validation.

Inverse Reinforcement Learning (IRL) aims to recover a reward function that explains expert demonstrations. Existing IRL methods typically rely on a bi-level optimization procedure that alternates between reward learning and policy optimization, leading to substantial computational burden and training instability. In this work, we introduce a different route that eliminates policy optimization entirely by leveraging diffusion policies. Our key insight is that a diffusion policy encodes the action-gradient structure of the optimal soft Q function, enabling reward learning to be cast as a sequence of value recovery problems, thereby allowing us to bypass reward-policy loops inherent in prior IRL methods. Specifically, our method proceeds in three stages: (I) recovering the optimal soft Q function via action-gradient matching and estimating the corresponding soft value function (LogSumExp of Q values) in a way inspired by Gumbel regression; (II) calibrating these soft values by inferring a state-dependent offset; (III) extracting the reward by enforcing Bellman consistency. This leads to Loop-Free Inverse Reinforcement Learning (LFIRL), a fully offline algorithm that operates in a simple, loop-free, and sequential manner. LFIRL is simple to implement and significantly improves training efficiency while maintaining strong reward recovery performance. Empirically, across Maze, Franka Kitchen, Adroit Hand Pen, and Push-T benchmarks, LFIRL achieves 2-3x speedup over the fastest baselines, while matching or surpassing state-of-the-art methods in reward recovery quality.

Large language model (LLM) agents could help security operations centers (SOCs) investigate advanced persistent threats (APTs) by turning weak leads into evidence for intrusion scoping and response. Yet success under one telemetry setting does not establish robustness to changes in log collection, retention, or sampling. We introduce APTInvestBench, a benchmark for evaluating cross-telemetry robustness in autonomous APT investigation. It comprises 370 cases across seven SOC-inspired conditions, derived from 56 report-informed attack reconstructions with 16.4 million log records. Agents investigate unverified leads and submit reports with record-level citations. Fixed action-level support requirements track sufficient evidence across available logs, query returns, and formal citations, separating telemetry limitations from acquisition and reporting gaps. Across eleven LLMs, agents acquire sufficient evidence for 44.3% of recoverable attack actions on average, while formal citations support only 25.0%. More importantly, aggregate coverage can conceal substantial instability: from Full to endpoint-only telemetry, coverage declines by only 1.6 percentage points, yet 35.5% of previously covered actions lose sufficient citation support despite remaining recoverable. Across four frameworks, such losses persist even when registered supporting records remain unchanged. APTInvestBench provides reusable investigation environments and diagnostic evaluation for identifying these gaps and developing more reliable defensive agents.

We prove two conjectures of Laugesen and Siudeja: the equilateral triangle uniquely maximizes the harmonic mean of the first two positive Neumann eigenvalues at fixed perimeter and their geometric mean at fixed area. The proof uses the two equilateral eigenfunctions and four projected polynomial corrections. Triangle symmetry reduces the associated variational matrices to two-by-two operations and two scalar invariants. Angular comparison and horizontal trial functions then yield the global inequalities. The same argument gives explicit quadratic stability bounds in terms of normalized side-length asymmetry.

We study the singular sets of real-valued solutions to periodic parabolic equations with rapidly oscillating coefficients. Under a nondegeneracy condition on the corrector matrix, we obtain a local parabolic $n$-dimensional Hausdorff measure bound and a codimension-two Minkowski estimate. The Minkowski estimate holds at every scale, including scales below the oscillation scale $\varepsilon$, with a constant independent of $\varepsilon$. The bounds depend on a coarse-scale ratio of space-time $L^2$ mass to terminal-slice $L^2$ mass. Independently of corrector nondegeneracy, we prove that this ratio controls doubling at all smaller scales uniformly in $\varepsilon$. The singular-set argument combines summable errors in truncated Gaussian approximations, comparison of caloric polynomials at different time centers, and a stopping cover with microscopic singular-set estimates.

Mastery-based education increasingly places the reporting of learning progress in students' hands, who record task completion on learning dashboards. The usefulness of such self-reports depends on how closely reported mastery corresponds to demonstrated competence. Most evidence on student self-assessment compares an overall self-rating with an overall examination score and therefore provides limited evidence about which tasks or which students account for the mismatch. This study examines first-year students' self-assessment against viva-demonstrated mastery at the level of individual tasks across a ladder of sixty programming tasks. The study draws on a large first-year programming course taught in 2023, involving 203 students and 12 examiners, in which every reported task was verified through an oral viva. Because a task entered the Viva only after it was reported, the design is one-sided and captures over-estimation but not under-estimation. Of 11,093 reported tasks, 10,885 (98.1%) were demonstrated, indicating a high degree of correspondence between self-report and demonstrated mastery. The remaining 208 overestimations were not evenly distributed. A small number of students accounted for most of the errors, and they occurred mainly on difficult tasks near the end of the task ladder rather than on higher-point tasks. This task-level analysis shows that high overall self-assessment accuracy can coexist with specific areas where reported and demonstrated mastery diverge. It also provides a practical basis for directing additional verification and formative feedback toward students and tasks where such divergence is more likely.

Regional seismic assessments often assign structural properties by building archetype, limiting their representation of building-specific responses. This study develops a modular physics-regularized framework for identifying equivalent single-degree-of-freedom Bouc–Wen models from seismic input–response histories. A differentiable forward solver enables joint estimation of hysteretic parameters and viscous damping through an objective combining response matching with independently weighted dynamic-equilibrium and hysteretic state-evolution residuals. Before regional application, synthetic systems and four-story steel-frame hybrid simulation data are used to evaluate identification and prediction under unseen inputs, showing improved parameter estimation relative to a genetic algorithm and improved predictive performance with physics regularization, respectively. The framework is then applied to a numerical benchmark of 14,280 buildings in Milpitas, California, using models identified from simulated responses to a smaller earthquake to predict responses under larger scenarios. With the inventory, ground-motion inputs, and damage-and-loss models held fixed, archetypal and response-informed portfolios yield distinct damage-fraction distributions and substantially different structural repair costs. For the earthquake scenario with moment magnitude $M_w=6.8$, the response-informed portfolio yields a mean structural repair cost nearly twice that of the archetypal portfolio, despite nearly identical mean damage fractions. These results highlight the influence of structural model assignment on regional damage and loss estimates and the importance of incorporating building-specific response information into regional assessment.

Text-video retrieval, which aims to bridge visual and textual modalities by learning a joint embedding space, has become a crucial task in multimodal intelligence. Despite extensive efforts to mitigate visual redundancy, previous methods typically rely on a single-aspect criterion to assess visual importance, overlooking the multifaceted spatiotemporal nature of video. In addition, encoding text into a single global embedding to align with videos compresses temporal events and spatial entities into a unified representation space, further aggravating cross-modal misalignment. To address these issues, we propose MMTI, a method that jointly mitigates visual redundancy and enables multi-grained text-video interaction to achieve accurate multi-grained semantic alignment. Specifically, a key feature selection (KFS) mechanism adaptively identifies and aggregates informative frames and patches by jointly evaluating multi-dimensional saliency and learnable importance scores, effectively compacting dense visual features and mitigating visual redundancy. Furthermore, our proposed multi-grained text-video interaction module (TVIM) employs a dynamic gating mechanism to decompose the text query into sentence, frame, and patch queries (SFP), enabling multi-grained text-video alignment. Complementary alignment at different granularities is thereby achieved. Extensive experiments on four standard benchmarks demonstrate that our method outperforms state-of-the-art methods.

Frontier models excel at many digital benchmarks, yet their ability to drive a real car, an everyday human skill, remains largely untested. We present DrivingBench, to our knowledge the first benchmark where general-purpose vision-language models must drive a real car. Through three tools, the models see camera frames from a Toyota Corolla and directly command its steering and velocity around a parking lot cone course at low speeds. The car may continue moving while the model thinks and new commands replace the currently running one, so inference latency is part of the task, testing the models' abilities to observe, act, monitor, recover, and complete a long-horizon objective under such constraints. We benchmark GPT-6 Astra, Claude Fable 5.1, GPT-5.6 Sol, and Grok 4.6 in vendor-native harnesses (Codex, Claude Code, Cursor) with up to three attempts each in one conversation; Astra is the only model to finish the course, on its second attempt, with no other attempt passing 50% of the course. Two of the four models improved materially across attempts with retained context. We also detail the design principles behind our action interface, and show how the tool output format and the framing of the task combined to determine whether models would drive at all or refuse. We release our harness, prompts, course map, and traces with video and telemetry for reproducibility.

We discuss a class of mixed characteristic rings defined analogously to Stanley-Reisner rings by replacing one variable with a uniformizing parameter for a discrete valuation ring. We adapt Hochster's formula for Tor and Ext modules, Hochster's formula for local cohomology, and Terai's criterion for Serre conditions to this new setting; one of the tools needed is cellular sheaf cohomology, for which we give a brief treatment.

Generative AI search and AI overviews are transforming access to information and news, renewing concerns that readers will encounter a narrower range of topics and have less in common. We examine these concerns via a randomized field experiment with 37,561 readers at The Washington Post. Both groups searched the same archive, but treatment readers also received AI answers with article citations above conventional results. Measuring consumption across displayed answers and opened articles, we find that AI search expands the reach of widely read topics and increases overlap in readers' topic consumption. At the same time, consumption becomes less concentrated and shifts toward less-popular topics, both within readers and across the audience. AI answers account for most of the increase in shared information, delivering it without requiring article clicks and broadening exposure beyond the articles readers open. Cited articles also contribute to the shift toward less-popular topics. Readers shift from conventional-result clicks and browsing toward cited articles and follow-up searches. More frequent searching offsets lower article consumption per search, producing a small increase in article consumption per reader. Total information consumption per minute also rises. Generative AI search can thus diversify collective attention while strengthening the information readers have in common.

The generation of certified random bits is an emerging near-term application of quantum computers. Potential applications, such as randomness beacons and CRS generation, require nearly uniform randomness, whose rate (the ratio of min-entropy to bitlength) is ~ 1. However, existing protocols for certified randomness either produce weakly random strings with rate o(1), or they require a trusted random seed. We show how to certify randomness with the optimal rate ~ 1 without requiring any trusted randomness from the verifier. Our protocol is secure unconditionally in the quantum random oracle model (QROM). This improves the result of Coladangelo et al., who certified weakly random strings unconditionally in the QROM, and answers a question posed by Aaronson and Hung. We also define and construct a proof of conditional min-entropy, which certifies optimal min-entropy even conditioned on an adversarially chosen transcript. We show an application of this primitive to randomness beacons, where each pulse should have high min-entropy conditioned on all previous messages.

We numerically investigate frequency-comb generation in an electrically pumped vertical-cavity surface-emitting laser coupled to two mechanical modes of a suspended mirror. Displacement-dependent cavity loss and radiation-pressure feedback couple the mechanical motion to the carrier and photon densities, producing harmonics and combination-frequency components in the photon-density spectrum. Mechanical quality factors primarily control the intensity of comb lines, whereas resonance frequencies determine their spectral distribution. The initial mechanical amplitudes influence the threshold current and relative spectral intensities, with the simultaneous excitation of both modes yielding a broader set of observable components. These results establish the dependence of the calculated comb structure on mechanical dissipation, resonance frequencies, and initial conditions in an active optomechanical cavity.

We introduce a quantum-echo Markov process for combinatorial optimization. Quantum dynamics based on quantum annealing (QA) or the quantum approximate optimization algorithm (QAOA) is used to engineer the transition kernel. Increasing the annealing time in QA or the number of layers in QAOA enables transitions to explore distant configurations while suppressing large energy changes. The Hamming-space delocalization originates from operator spreading, whereas the energy-space localization arises from a dynamically generated correlation. For optimization, we propose quantum-echo local optimization with and without greedy descent. We find that its performance is governed by the interplay between Hamming-space nonlocality and energy-space locality, and that excessive energy-space localization can degrade optimization performance. Incorporating greedy descent substantially improves the performance, highlighting the complementary roles of quantum dynamics for exploration and greedy descent for exploitation. These results establish quantum-echo dynamics as a framework for engineering structured transition kernels and provide a route to using finite-resource quantum many-body dynamics as a computational primitive for iterative optimization.

In this work, the structural stability, and the electronic and magnetic properties of a novel CeN monolayer derived from the (111) surface of bulk rock salt CeN are investigated using first-principles calculations. Dimensional reduction produces substantial changes in the local atomic environment, decreasing the Ce-N coordination from sixfold in the bulk to threefold in the monolayer and shortening the Ce-N bond length, while preserving an essentially planar hexagonal structure. Moreover, the electronic and magnetic properties are significantly modified, which may be useful for practical applications. Both ferromagnetic (FM) and antiferromagnetic (AFM) configurations were investigated and found to be strongly bound, with the FM state slightly lower in energy. The magnetic moments are predominantly localized on the Ce atoms and mainly originated from the Ce 4f states. The electronic structure also exhibits a pronounced dependence on magnetic ordering, highlighting the connection between localized 4f states and the spin-dependent electronic properties of the system. In addition, the calculated elastic properties confirm the mechanical stability of the monolayer, while phonon calculations support its dynamical stability. Ab initio molecular dynamics simulations further indicate that the two-dimensional structure preserves its integrity under finite-temperature conditions. Overall, these results demonstrate that dimensional reduction significantly modifies the structural, electronic, and magnetic behavior of CeN without compromising the stability of the resulting two-dimensional lattice. The combination of structural robustness, localized Ce 4f magnetism, and spin-dependent electronic properties make the CeN monolayer an interesting system for further exploration in two-dimensional magnetism and spin-dependent applications.

We show that assuming the existence of homomorphic encryption, parallel repetition of all interactive arguments (after being run under homomorphic encryption) reduces the soundness error at a tight exponential rate even in the post-quantum setting. Moreover, we generalize this result to hold for threshold verifiers, where the parallel repeated verifier accepts if and only if at least $t$ of the executions are accepted (for some threshold $t$). Prior to this work, these results were known only when the cheating prover was assumed to be classical, and it was not known how to achieve tight bounds. As a corollary, we construct the first constant-round succinct argument for $\mathsf{QMA}$ with negligible completeness and soundness errors assuming only the existence of quantum homomorphic encryption.

A universal empirical coupled channel model, ECC2, is proposed for the systematic description of capture cross sections in heavy-ion fusion reactions. The model combines a modified Broglia–Winther potential, an effective barrier curvature, and a deformation-dependent barrier distribution width. The most probable barrier heights predicted by ECC2 are in good agreement with the extracted values based on measured excitation functions for 367 systems, and the root-mean-square (rms) deviation is 1.39 MeV. ECC2 is applied to 25 fusion reactions ranging from $^{16}$O+$^{197}$Au to $^{54}$Cr+$^{248}$Cm. The predicted capture excitation functions are consistent with the EBD2.2 predictions for all systems and agree well with the available experimental data.

In this work, we propose a defect-aware parallel atom reloading protocol that reloads atoms based on defect information to increase the atom filling rate. The proposed protocol is useful for neutral-atom quantum computers, where reloading atoms into the defects, or atom-loss positions, is mandatory for continuous operation against atom loss. Our protocol takes advantage of both parallelism and defect awareness to improve atom-reloading efficiency. The proposed protocol consists of the coherent atom reloading operation and the defect-aware planner. The former replenishes the defects with atoms preserving the quantum information coherently. This coherence allows us to adaptively choose the reloading strategy based on the current defect configuration. The latter determines the atom reloading pattern to efficiently replenish the defects. In numerical experiments assuming a low-noise environment, we confirmed that the average atom filling rate improved from 98.61% to 99.94% for a 36 x 90 atom array, matching the filling rate obtained with a mixed integer programming solver to within 0.001 percentage points while planning each reload in under 0.1 ms. These results demonstrate that defect-aware parallel reloading can achieve and sustain high atom filling rates while the performance of the defect-aware planner meets the real-time requirements of neutral-atom quantum computers.