Papers

World models enable large language model (LLM) agents to anticipate action effects and make informed decisions. For generalization, it is desirable for such world models to transfer efficiently to new environments with similar interaction structure and underlying dynamics. Recent approaches encode world dynamics as symbolic rules or executable programs, whose compositional structure in principle affords transfer, but still organize such knowledge around environment-specific objects. To address this limitation, we introduce RoleWorld, a symbolic world modeling approach which uses reusable prototypes and environment-specific parameter values as functional role abstractions to represent how objects behave in interactions. From interaction trajectories, a teacher LLM extracts role prototypes and executable transition rules defined over role instances, together with a skill that guides the actor LLM to use role information in decision making. In a target world, without the teacher, the actor uses this skill together with its current role assignments and parameter estimates, and actively tests a role instance hypothesis through interaction when its uncertainty affects the current decision. Feedback from each action refines role assignments and estimates parameter values online, with no LLM parameter updates and the transition rules held fixed. Experiments on MARS benchmark show that RoleWorld enables robust generalization across environments with significantly altered object behavior, suggesting that functional role abstractions can serve as an effective component for knowledge transfer in world modeling. Project Page: https://future-item.github.io/RoleWorld.

Power sampling improves large language model (LLM) reasoning at test time by sampling from a sharpened version of the base model's distribution. Its sampler, PowerMH, is slow because its Metropolis–Hastings (MH) updates are sequential: each accept/reject decision sets the state from which the next proposal is generated. Prior work reduces this cost by changing the sampler or the model (e.g., sequential Monte Carlo, approximate lookahead, distillation). We propose Predictive SubTree Prefetching (PreSTO), which accelerates PowerMH by prefetching future proposals and provably leaves its MH transition kernel unchanged. PreSTO views future MH outcomes as a tree of accept/reject decisions: in each model call, it generates the proposals of a prefetched subtree of this tree in one batch, then follows the realized path with the original MH acceptance rule. We also observe that MH reaches only a sparse, nearly unary part of each prefetched subtree, we further design a predictive traversal rule that chooses the subtree to maximize the predicted number of completed transitions. Across seven datasets and five base LLMs, PreSTO achieves median speedups of $1.71\times$, $1.91\times$, and $2.33\times$ on PowerMH and its follow-ups EntropyCut and MultiTryMH, respectively; on the first two, it makes a median of $3.9\times$ fewer model calls, each only $2.1\times$ longer. On LCB V6 with Qwen3.5-9B, the predictive traversal rule is a further $1.25\times$–$1.37\times$ faster than the default breadth-first traversal.

With the rapid development of blockchain technology, numerous heterogeneous blockchains have emerged, each featuring distinct architectures, consensus mechanisms, and application ecosystems. Cross-chain bridges have consequently become a critical infrastructure component, typically relying on smart contracts to facilitate message passing and asset transfers across heterogeneous blockchains. Unlike single-chain contracts, bridge contracts introduce additional verification logic to validate external-chain data, such as state proofs, message sender identities, and signatures. However, correctly implementing such logic is challenging. We refer to vulnerabilities that violate the security guarantees of cross-chain verification as Cross-chain Verification Vulnerabilities (CVVs). To characterize CVVs, we conduct the first systematic empirical study of this vulnerability class. By analyzing 588 security audit reports from 36 real-world cross-chain bridge projects, we derive 11 CVV patterns and organize them into 5 high-level categories according to the affected verification properties. Following this taxonomy, we construct a CVV dataset containing 830 vulnerability instances across 786 contract files. The dataset establishes mappings among audit reports, audited source-code versions, affected source files, and CVV-pattern labels. We then assess existing detection approaches. Current tools provide limited documented coverage of CVV patterns, while only two of four evaluated LLMs outperform an all-pattern constant baseline of 29.3% Macro-F1; the best reaches 51.2%. Together, the limited tool coverage and inconsistent LLM gains motivate detection methods tailored to cross-chain verification semantics.

The imbibition of milk into a biscuit provides an example of a broader class of processes in which a liquid invades a porous material, dissolving components of the solid and thereby modifying its own rheology. This coupling between imbibition, dissolution, and evolving fluid rheology also occurs in food processing, filtration, and reactive porous media. While the resulting changes in fluid viscosity are known to affect imbibition, a quantitative description of how this evolving rheology determines the imbibition dynamics is still lacking. Here, we establish such a description for dissolution-driven fluid thickening. For a capillary tube in the well-mixed limit, we derive a generalized Washburn law showing that the evolving viscosity can be incorporated through a rescaled time given by the accumulated inverse viscosity. This predicts a collapse of the imbibition dynamics for different dissolution rates, which we confirm numerically using coupled free-surface and solute-transport simulations. When diffusion is slower, concentration and viscosity become spatially heterogeneous, resulting in a weaker retardation of the advancing liquid than predicted by the well-mixed theory. We then consider disordered porous media and quantify how dissolution-driven thickening and variations in pore geometry affect the dynamics. Our results extend the classical Washburn description to imbibition with evolving fluid properties and provide a quantitative basis for understanding fluid uptake when interactions with the solid progressively modify the rheology of the invading liquid.

We investigate restrictions on the dimensions and coordinate fields of real Weyl-Heisenberg fiducial vectors. Our main result is that such vectors can exist only in dimensions $d\equiv 3\pmod{4}$. This result is motivated by evidence showing that real Weyl-Heisenberg fiducial vectors can exist only in dimensions $d = 4m^2 + 3, m \in \mathbb{Z}_{\geq 0}$. We further restrict the possible dimensions when the coordinate field is a totally real number field, an even or odd degree extension of $\mathbb{Q}$, a real quadratic field, or a maximal real subfield of a cyclotomic field. Lastly, we establish non-existence over odd-degree number fields under either positivity of all non-trivial cyclic-shift overlaps or a conjectural containment of a distinguished quadratic field in the SIC field.

We compare the Nordic Grid Development Perspective 2025 (NGDP) with a least-cost transition pathway for the Nordic power system obtained with PyPSA-Eur for the period 2025-2050 under a net-zero CO$_2$ constraint. The model co-optimises generation, storage, and transmission expansion with hourly system operation while constraining the Nordic generation mix to remain consistent with NGDP milestone ranges. Overall, the optimised capacity pathways reproduce the broad NGDP expansion of wind and solar power, with Denmark emerging as the main locus of offshore wind and solar deployment, Finland as a major site for onshore wind expansion, and Norway and Sweden relying more strongly on existing hydropower and nuclear capacity. The main discrepancy between the two pathways does not lie in the broad build-out of variable renewables, but in the timing and role of hydrogen. Whereas the NGDP treats hydrogen as an early and steadily increasing component of industrial demand and system development already from 2030, PyPSA-Eur delays substantial hydrogen deployment until 2045-2050, when it becomes cost-effective primarily as a flexibility and seasonal storage option in a strongly variable-renewable-energy-dominated system. This contrast suggests that early hydrogen build-out in the Nordics is not a robust outcome of cost optimisation alone, but depends strongly on assumptions regarding industrial demand growth, sector coupling, and anticipatory infrastructure planning. We therefore argue that hydrogen remains the principal strategic uncertainty in the Nordic transition, with the NGDP representing a more anticipatory planning perspective and PyPSA-Eur a more conservative least-cost timing of deployment.

Terminal agents are moving beyond local operations toward completing real-world professional tasks. Short trajectories can demonstrate commands and tool calls, but they do not fully show how agents accumulate evidence through exploration, use it to inform judgments, continually adjust strategies based on feedback, and turn intermediate results into final deliverables. High-quality long-horizon trajectories capture these agentic behavioral patterns and their connections within a single problem-solving process, providing richer experience for sustained autonomous work. However, environment synthesis may complete or simplify the key work underlying these patterns in advance and thus remove the need for exploration, implementation, and verification that the agent was meant to perform. To address this, we introduce TerminalHorizon, a fully automated data synthesis engine for long-horizon terminal agents. Starting from real-world professional work, it uses independent execution to identify key work omitted during construction and uses the resulting evidence to guide environment reconstruction. To extend these patterns across stages, it progressively expands tasks under the same goal, introducing new requirements that build on prior results and call for further exploration and improvement. Using this engine, we construct 1.5K task environments and perform supervised fine-tuning of Qwen3.5-27B on the resulting dataset to obtain TerminalHorizon-27B. We achieve substantial performance gains on all 12 benchmarks spanning coding-agent tasks, agentic tasks in real-world production settings, and general reasoning. Multiple RSI case studies, each with a 12-hour research budget, further demonstrate the model's ability to continue experimenting beyond an initial solution, improve its methods based on feedback, and retain effective improvements in its final deliverables.

Rare decays of $B$ hadrons, such as $b\rightarrow s\ell \ell$ processes, provide great avenues to search for the influence of new physics (NP). They are highly suppressed in the Standard Model (SM) and therefore sensitive to NP contributions. In recent years, experimental measurements have shown tension with SM predictions. It is still up for debate whether these stem from genuine NP or an incomplete understanding of the theory. These proceedings focus on the recent efforts of the LHCb collaboration to advance this field. Some of the results covered show that tensions in these processes, like in $B^0 \rightarrow K^{*0}\, \mu^+ \mu^-$, still persist upon higher precision. Others aim to shed some light on the theoretical picture through amplitude analyses of decays like $B^0 \rightarrow K^{*0}\, \mu^+ \mu^-$ and $B^+ \rightarrow K^+ \, \mu^+ \mu^-$. We also present two innovative measurements: the branching fraction of $B^0 \rightarrow K^+ \pi^- \mu^+ \mu^-$ at large dihadron mass; and the first hadron collider measurement of the lepton flavour universality ratio of $B^0 \rightarrow K^{*0} \, \ell^+\ell^-$ at high $q^2$ (dilepton mass squared).

Scaling robot manipulation policies requires large amounts of diverse, action-labeled data, yet collecting such data in the real world remains expensive and difficult to scale. Existing methods either depend on human teleoperation, rely on simulation with limited visual realism, or use human videos that lack embodiment-specific annotations. In this paper, we introduce Grounded-Dreamer, a framework for synthesizing physically grounded robot manipulation data with video world models while avoiding human teleoperation demonstrations. Grounded-Dreamer combines the visual realism of generative video models with simulator physics engines. Starting from a small set of real images depicting robot task initial states, our method automatically reconstructs task-relevant objects and camera geometry, aligns them in simulation, and collects physically valid trajectories. These trajectories are used to post-train both a video world model and an inverse dynamics model, enabling the generation of visually realistic robot manipulation videos and corresponding robot actions from real images. To filter physically implausible rollouts, Grounded-Dreamer replays generated trajectories in the reconstructed simulation and retains only those that successfully complete the tasks. The real-world robot policies are trained solely on the grounded synthetic trajectories without any human demonstrations. Empirical results in real-world tasks demonstrate that Grounded-Dreamer can synthesize physically grounded data and substantially facilitate policy learning without requiring human teleoperation effort.

After removing suitable thin subsets, we count rational points of bounded height on some hypersurfaces of small bidegree in $\mathbb{P}^3\times\mathbb{P}^3$ and in $\mathbb{P}^3\times\mathbb{P}^2$. One of the ingredients is a new estimate for the number of rational points of bounded height on diagonal quadratic surfaces which improves as the coefficients grow.

This paper studies entropy regularization for the discrete-time dynamic mean–variance portfolio problem with no-shorting constraints. In the unconstrained setting, entropy regularization can be introduced without changing the basic quadratic structure of the dynamic program. Under no-shorting, this property is lost: a standard fixed-temperature formulation generally preserves convexity but no longer preserves the homogeneous Bellman structure and the associated piecewise-linear feedback form. We introduce a state-dependent temperature that is matched to the homogeneity of the original mean–variance problem. This restores the piecewise-quadratic value function and the finite-dimensional Bellman recursion. The resulting formulation, however, makes convexity depend on the continuation coefficients generated by the soft recursion. We show that negative continuation coefficients lead to loss of convexity and prove that a common sufficiently small temperature preserves the Bellman structure and convexity over the entire finite horizon. The original no-shorting solution is recovered as the temperature vanishes. We also develop a sample-based implementation for unknown return distributions. Numerical and empirical results illustrate both the structural advantage of the state-dependent temperature and the practical relevance of the convexity-preserving temperature restriction.

Executing large network configuration changes, called migrations, is slow and error-prone today; operators must manually devise a safe order for updating devices that does not disrupt traffic and meets operational intents (e.g., low total time). We develop Smoothie, a versatile migration planning framework that, unlike existing tools, can support diverse operational intents. Inspired by automated planning, it achieves versatility by decoupling the encoding of network safety and intents from the search for a viable plan. Smoothie uses symbolic simulation for general and scalable safety encoding, after which it can use a range of planners, including existing off-the-shelf planners and a new one we developed for optimized parallel plans. We demonstrate Smoothie's versatility and benchmark its performance using diverse migration scenarios on many topologies. It finds both sequential plans and parallel plans, and the parallel ones compress up to thousands of configuration updates into no more than 3 steps. Its versatility does not come at the expense of performance: finishing most planning tasks in less than 1 hour.

We study the Hausdorff dimension of attracting sets of three-dimensional flows whose one-dimensional quotient dynamics is non-uniformly expanding. Our main assumption is that the quotient map admits a full-branch uniformly expanding Gibbs–Markov induced map. We first prove that every such attractor contains a hyperbolic horseshoe of Hausdorff dimension strictly larger than two. As a consequence, after an arbitrarily small $C^2$ perturbation, the associated dynamical Lagrange and Markov spectra have robustly non-empty interior for an open and dense set of $C^1$-observables. We then establish intermediate dimension properties. Every admissible intermediate Hausdorff dimension is realized by an ergodic invariant measure supported on the attractor. Moreover, every intermediate stable and unstable dimension is realized by a strictly ergodic hyperbolic subsystem. The approach extends the dimension theory previously developed for geometric Lorenz attractors to more general non-uniformly hyperbolic attractors including singular-hyperbolic attractors, Rovella attractors, and broad classes of skew-product suspension flows modeled by intermittent and Benedicks–Carleson maps.

Currently available methods for processing atmospheric wave disturbances are often based on integral wavelet transform techniques, which efficiently estimate spectral data but are fundamentally limited by the uncertainty principle. To address this limitation, this work proposes a novel morphological approach based on partitioning spatial disturbance profiles into separate, well-defined wave segments. By focusing exclusively on the local waveform morphology of a central half-wave and its two adjacent out-of-phase counterparts, and taking the modified Morlet wavelet as a primary example of a model wave packet, we successfully reconstruct the complete set of local wave parameters directly from the amplitudes and positions of these three dominant half-waves. Consequently, the theory demonstrates good agreement with observations, confirming the high accuracy of the proposed waveform morphology approach in reconstructing the localized structure of atmospheric disturbances. When applied to evaluating the phase shift between different wave field components, the method reveals that this shift is a weighted combination of the classical monochromatic phase difference and a morphological one determined by the internal symmetry of the underlying wave packets. Although the classical factor yields the dominant contribution, the symmetry-related factor provides a moderate correction to it. The applicability and accuracy of the approach are verified using observational data gathered over several orbits of the low-Earth-orbit satellite Dynamic Explorer 2 as it passed over Earth's polar caps. The obtained results confirm that the observed disturbances correspond to medium-scale acoustic-gravity waves propagating at a small angle to the horizontal plane, demonstrating the broader methodological potential of the proposed technique for analyzing a wide range of wave phenomena.

Reactive collision avoidance in unknown, cluttered environments must perceive obstacles and correct motion within milliseconds. Existing methods rely on maps for high-rate collision queries, but their acceleration structures require pre- processing. That cost is amortized in long-horizon planning, where a map serves millions of queries; reactive planning, however, is a near-field, short-horizon problem with potentially only hundreds of queries per measurement, where building a map may not pay off. We propose a mapless, Graphics Processing Unit (GPU)-accelerated Euclidean Signed Distance Field (ESDF) query with negligible pre-processing that operates directly on posed depth images, with resolution set by pooling the image rather than by a voxel size. On an embedded GPU it is simultaneously more accurate and lower in total latency than every voxel-grid configuration we measured between 2 cm and 20 cm, at one to two orders of magnitude less memory at matched resolution. With our query integrated into a Model Predictive Controller (MPC), simulated runs in laser-scanned forests at up to 9 m/s collide in 0.4 % through dense branches, against 5.9 % with the best voxel grid and 5.5 % with the best mapless baseline at our resolution, whose failures we attribute to voxel resolution and total latency respectively. We deploy the system onboard at two forest sites and fly at commanded speeds of up to 4 m/s.

Reinforcement learning (RL) for search agents often suffers from evidence fabrication, a failure mode where agents bypass external information ($E_{ext}$) and fabricate an incorrect answer. We argue that this fabrication arises from an over-reliance on unreliable internal knowledge ($E_{int}$) as shortcuts due to the inherent opacity of reasoning trajectories. To address this, we propose SOAR (SOurce-Aware Reinforcement learning), a paradigm that transforms monolithic reasoning into source-explicit trajectories. SOAR (i) disentangles $E_{ext}$ and $E_{int}$ into parallel, observable streams via a Source-Aware Workflow, and (ii) incorporates Source-Consistent Reward to guide the policy evolution. By making information-source attribution explicit, SOAR enables precise credit assignment between different knowledge streams. Evaluations show that SOAR outperforms baselines on 3B and 7B backbones by 7.3% and 5.4%, while reducing evidence fabrication by 74.8%. These confirm that SOAR significantly improves performance on knowledge-intensive benchmarks while effectively mitigating evidence fabrication, fostering more faithful and grounded reasoning with enhanced robustness.

Studies of conversational agents rely on interviews, surveys, lab tasks, or taxonomies. Whether anthropomorphic naming splits those patterns, and whether replies should change, lack field evidence. We report a field study of a desktop assistant (May-August 2026; n=257). Stance is an anthropomorphic assistant name (118 users; 45.9%). Behavior comes from a 128-unit sparse-autoencoder codebook, identity tags, and opening lexicons. Ranks differ by stance (Spearman rho=+0.249 vs. null mean +0.382; p=.0419). Teacher and tutoring identities covary with renaming, an introverted task-only orientation against it; renamers greet more and command less; confiding shows the codebook's largest gap. We revise the agent: greet and receive affect on relational openings, execute directly on tasking. In 1,744 judgments (n=129), renamers chose the revised reply far more often than default-name users on being treated as a person and having affect received, less robustly on willingness to continue, and neither judged it worse on task interference.

The degree of oxygen vacancy in A-site-doped layered double perovskite oxides can lead to anion vacancy ordering, which subsequently results in distinct superstructures. We have studied oxygen deficient double perovskite oxide, NdBaCo$2$O$_{5+\delta}$ ($\delta \sim$ 0.65), utilising X-ray absorption near edge structure (XANES) spectroscopy complemented by multiple scattering and first-principles calculations based on density functional theory (DFT+U). The present results establish a direct correlation between oxygen-vacancy distribution, magnetic inhomogeneity, and charge transport in layered cobaltates. Our investigations on the Co K, Ba, and Nd $L_{\mathrm{III}}$ edges suggest that the near-edge spectra, based on the number and shape of the spectral features, indicate the value of $\delta$ for the compound under study lies closer to 0.75. The analysis of the Ba $L_{\mathrm{III}}$ edge spectrum confirms the +2 oxidation state for Ba ions. The Co K-edge near-edge spectra suggest the possibility of oxygen vacancies arising at the basal sites in addition to the apical sites. This is counterintuitive to the common understanding that oxygen vacancies usually occur at the apical sites, i.e., in the lanthanide (Nd) layer in these compounds. Such a distribution of oxygen vacancies naturally explains the Griffith phase and the conduction mechanism of variable range hopping observed in the compound. The vacancy distribution, when at the optimum level, can also bring about easier oxygen ion mobility and a more sensitive and faster oxygen sensor. We believe that our results will help in tailoring oxygen-ion mobility and enhancing oxygen-sensing performance.

Multimodal Large Language Models (MLLMs) generally fall short of expectations in Fine-Grained Visual Recognition (FGVR), although they exhibit promising FGVR foundations, i.e., visual perception and broad category knowledge. Most existing studies seek to remedy this deficiency by adopting domain-specific post-training, at the risk of compromising generalization to unseen domains. Different from these efforts, in this paper, we revisit the FGVR mechanism underlying MLLMs from both the empirical and theoretical perspectives and uncover an inference-time phenomenon termed FGVR Laziness. In brief, FGVR Laziness refers to the insufficient association between visual cues and category concepts under certain candidate category lists and recognition guidance, leaving the FGVR foundations of MLLMs largely under-utilized. To mitigate FGVR Laziness, we propose Laziness-resIstant Visual rEcognition (LIVE), a training-free framework which jointly refines candidate lists and recognition guidance to elicit the latent FGVR capability of MLLMs. Specifically, LIVE first performs query-adaptive candidate inference to derive a compact yet reliable candidate list. Meanwhile, LIVE constructs category-specific FGVR skills by mining the intra-category characteristics and inter-category differences. Finally, LIVE selectively invokes detailed skill guidance according to recognition uncertainty and category confusion, shifting MLLM-based FGVR from a cursory glance to a focus on associating discriminative visual cues with category concepts. Extensive experiments on 13 benchmarks against 24 state-of-the-art methods spanning fine-grained category recognition, action recognition, and general category recognition, verify the effectiveness and generalizability of LIVE across different MLLM backbones. The code will be released upon acceptance.

Nitrogen-vacancy (NV) centers in diamond are promising quantum sensors for AC magnetic fields. However, conventional dynamical-decoupling-based magnetometry becomes technically demanding at high frequencies because of the requirement for increasingly short and high-power microwave pulses. This paper describes an AC magnetometry technique based on continuous-wave optically detected magnetic resonance (CW-ODMR) measurements, which enables the detection of signals in the hundred-MHz range. By applying a strong static magnetic field transverse to the NV axis, the energy separation between the relevant electron-spin states was tuned and resonantly coupled with the target radiofrequency (RF) magnetic field, thereby forming RF-dressed states. Unlike pulsed dynamical-decoupling schemes, this approach does not require high-power microwave pulse sequences. The key feature of our method is that the detectable frequency can be tuned by controlling the bias magnetic field strength. Using this methodology, we have experimentally achieved the frequency-tunable detection of AC magnetic fields from 14.4-632 MHz. The sensitivity was 0.558 uT/Hz^(0.5) at 14.4 MHz and 24.4 uT/Hz^(0.5) at 632 MHz. We characterized the field-dependent ODMR contrast and linewidth, and found that their degradation at large transverse fields is a important factor limiting the high-frequency sensitivity. These results establish a simple CW-ODMR approach for widely tunable RF magnetic-field sensing without high-power pulsed spin control.

Open coding agents are commonly trained through supervised fine-tuning on interaction trajectories whose final patches pass the benchmark test suite. However, this outcome-level filtering criterion ignores process quality: trajectories that eventually succeed may still contain inefficient interaction patterns such as repetitive file inspection, unstable tool usage, or delayed modification behavior, all of which become part of the imitation target. We propose TQS (Trajectory Quality Score), a structural scoring framework for coding-agent trajectories that operates directly on trajectory structure, requires no LLM judge, and runs in linear time. TQS characterizes trajectory quality through structural signals organized around structured convergence: information focus, intent–action alignment, and policy dynamics. Using 155,144 successful trajectories from a public open-license corpus, we rank trajectories by TQS and fine-tune coding agents on filtered subsets. Models trained on top-ranked trajectories consistently outperform those trained on random or bottom-ranked subsets on SWE-Bench-Verified. Our results suggest that process-level trajectory quality is an important factor in coding-agent supervision and that structural trajectory filtering provides an effective and scalable alternative to LLM-based data selection.

We investigate the time-asymptotic stability of composite waves consisting of a non-classical Liu shock and a rarefaction wave for non-convex viscoelastic system. By choosing delicate time-dependent shift functions to the viscous shock and suitable weight functions for the relative entropy, we succeed in proving the time-asymptotic stability of the composite of a viscous Liu shock and a rarefaction wave from two different characteristic families for the one-dimensional viscoelastic system with a cubic non-convex stress-strain relation by $a$-contraction method. Here the non-classical Liu shock can be either degenerate or non-degenerate, which can be seen as the system version of the scalar Oleinik shock. Note that both the shift and the weight functions are chosen quite differently in the two cases of nearly degenerate and fully non-degenerate. The paper represents the first time-asymptotic stability result for the composite waves consisting of both non-classical Liu shock and rarefaction wave from two different families for the non-convex physical system.

Three-dimensional Einstein gravity has no local degrees of freedom, yet retains rich global and black-hole physics. Adding a scalar normally spoils this. We show that closure of three-dimensional Horndeski gravity under disformal field redefinitions selects a square-root curvature interaction. A nonlinear analysis finds no local bulk modes on regular branches, while an independent Hamiltonian analysis confirms the result on the timelike-gradient branch. This extends the no-local-bulk-dynamics paradigm beyond pure Einstein gravity and motivates studying its boundary and global sectors.

Rydberg excitons in atomically thin semiconductors offer a sensitive view of dielectric screening and electronic structure. As their spectra become increasingly well resolved, a central question is how much material information an effective Hamiltonian can predict beyond the states used to determine it. Using the 1s, 2p and 2s levels to determine a scalar Rytova-Keldysh Hamiltonian, we test its transfer to the excluded 3p state. The transfer deviations have opposite signs: the observed 3p lies 7.38 meV below the prediction in MoSe2 and 5.58 meV above it in WS2; the opposite signs persist across the complete reported +/-2-sigma peak-error boxes. In a separate high-field-calibrated benchmark, RK removes 66.8% of the orbital error of a pure two-dimensional Coulomb reference, making the remaining state-selective deviations particularly informative. After separating gap and common-scale contributions, the residuals reveal distinct material-dependent orbital patterns and constrain simple multichannel descriptions. For WS2, the measured 2p/3p curvatures yield a parameter-fixed prediction, Q4p=+4.08 meV, for a future same-platform measurement. Together these results show how a well-performing spectral model can be tested for transferable physical information, with the next unmeasured state providing an independent experimental check.

In this work, we revisit the classic Les Misérables benchmark network from the perspective of social communication. Using the 1998 film adaptation of the novel, we construct the publicly available Les Misérables Dialogue temporal text network, which explicitly distinguishes speakers, addressees, and other recipients of each utterance, along with its content and timing. Using the ten social dimensions model, we also derive multilayer network representations of the network, in which layers correspond to distinct relational dimensions. Beyond the network dataset, this work offers a general methodology for building similarly rich communication datasets from similar narrative sources.

The physical origin of foreshocks remains one of the major unresolved problems in statistical seismology. In bottom-up (BU) triggering models, foreshocks are interpreted as ordinary earthquakes that stochastically trigger a larger event through cascading interactions, as in the Epidemic-Type Aftershock Sequence (ETAS) model. In contrast, top-down (TD) models interpret foreshocks as manifestations of a preparatory process preceding the mainshock. This ambiguity has important implications not only for earthquake forecasting, but also for declustering itself, since the inferred number of foreshocks may depend on the assumed physical mechanism governing seismicity. Here we investigate the detectability of foreshocks using synthetic catalogues representing two limiting physical scenarios: purely bottom-up triggering and mixed preparatory-triggering dynamics. These cases are implemented through the ETAS and ETAFS models, respectively. We compare the classical nearest-neighbor (NN) declustering metric with a reverse nearest-neighbor metric (rNN) introduced here. While the standard NN metric scales the correlation metric to the magnitude of the earlier event, rNN scales it with the magnitude of the later event. We show that rNN is mathematically equivalent to applying the standard nearest-neighbor metric to the time-reversed catalog, establishing a direct symmetry between the two approaches. Our simulations reveal [...]

Belief propagation (BP) is widely used for inference on probabilistic graphical models but often suffers from non-convergence and poor accuracy on loopy factor graphs, yielding inexact variance estimates in the Gaussian case. Generalized BP (GBP) improves accuracy but relies on manual design or application-specific heuristics for region-graph construction. We propose extended neighborhood message passing (ENMP), which automatically constructs extended neighborhoods of variable nodes from the bipartite factor graph topology, capturing loop-induced correlations without manual region design. The framework applies to general distributions on factor graphs with a tunable accuracy-complexity trade-off. In the exact regime, message updates converge to the unique fixed point of the ENMP equations in finitely many iterations, yielding exact marginals; in the approximate regime, ENMP provides approximate marginals. For Gaussian factor graphs, we derive closed-form update rules enabling exact variance estimation in the exact regime, addressing the long-standing inaccurate variance estimates of loopy Gaussian BP. We evaluate ENMP on densely loopy Gaussian factor graphs from power-grid benchmarks and on a discrete Ising factor graph. Results show accuracy improves with neighborhood size, with exact marginals achieved on several systems in both settings. On larger Gaussian systems, ENMP outperforms regularized Gaussian BP (sGaBP-m) in both mean and variance estimation accuracy.

Large language models (LLMs) increasingly generate structured layouts and executable scene programs from natural-language instructions. However, executability does not imply engineering feasibility: an industrial scene program may execute successfully while its final state still violates coupled boundary, collision, clearance, aisle, or structural constraints. We identify this mismatch as the executability-engineering-feasibility gap, where locally plausible placements can remain globally infeasible because corrections to one device may introduce violations elsewhere. To make this gap measurable in a controlled setting, we construct an execution-grounded industrial scene-program dataset with 1,749 constructible scenarios, complete executable programs, and 94 final-state geometric checks per scenario. We further propose VIGOR, which combines supervised scene-program generation with execution-grounded verification and staged geometric refinement. Supervised fine-tuning establishes complete program generation, local projection resolves independently correctable violations, and joint conflict-graph repair coordinates residual multi-object conflicts while preserving structural relations. Experiments show that VIGOR achieves 96.86% Complete Feasibility and an Overall score of 91.20, outperforming the strongest evaluated baseline by 10.42 points in Overall. These results show that, under controlled industrial layout conditions, reliable scene-program generation requires not only executable programs, but also explicit verification and coordinated resolution of coupled engineering constraints.

LLM-based multi-agent systems (MAS) have emerged as a promising paradigm for enhancing complex reasoning through role specialization and collaboration, yet a key challenge remains: how can collaboration be adapted to a backbone's competence on the current query? This problem matters because more collaboration is not always better–richer coordination may help on difficult cases, but it also incurs additional token cost and redundant interactions. We propose a two-stage competence-aware framework for automatic MAS orchestration. It first models the LLM backbone's task-conditioned competence for the current query, and then uses this signal to adapt role activation and the communication graph. Across six benchmarks, our method achieves an average score of 93.81%. Experiments with six LLM backbones of varying competence levels further show that our method adjusts collaboration according to the inferred competence: for higher-competence backbones, it improves average accuracy from 87.22 to 89.97 while reducing token cost by 36.26%; for lower-competence backbones, it yields a further average gain of 4.58 points over existing methods. The same trend also holds on unseen backbones.

Image-based 3D reconstruction provides a scalable representation for building inspection and maintenance. However, photorealistic reconstruction alone does not fully provide the geometric measurability, semantic structure, and interactive information access required for engineering applications. This study proposes a measurable semantic 3D Gaussian Splatting (3DGS) framework with region and depth supervision for interactive building inspection from multi-view imagery. Mask-based region supervision suppresses irrelevant backgrounds and focuses optimization on target regions. Point-wise depth supervision guides Gaussian centers toward physical surfaces using reference geometry derived from multi-view stereo (MVS) point clouds, providing geometric anchors for measuring physical dimensions and areas. A semantic Gaussian field associates learnable class features with individual Gaussian primitives, enabling semantic rendering and component-level inspection. These capabilities are operationalized through AEC-SplatTwin, an interactive building inspection platform supporting real-time visualization, semantic analysis, quantitative measurement, and multimodal large language model (MLLM)-assisted interaction. On the synthetic UAV-inspection QuakeCity and smartphone-captured HKUPB datasets, the framework reduces Gaussian-center depth MAE by 58.1% and 86.8%, while its semantic field achieves mIoU values of 0.8529 and 0.8642. On HKUPB, the mean relative errors for height, width, and area measurements are 0.33%, 1.07%, and 0.60%. These results show that the proposed framework extends 3DGS beyond visual reconstruction into a measurable, semantically structured, and interactive engineering representation for building inspection.