Mesoscale and Nanoscale Physics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Friday 4 September
Multipole moments provide a unified symmetry language for describing electronic degrees of freedom and their associated physical responses. Although active multipoles have been systematically classified for crystallographic and magnetic point groups, their classification in spin point groups, which naturally describe magnetic systems in the absence of spin-orbit coupling, has remained unexplored. In this work, we present a complete classification of active orbital and spin multipoles for all 32 nonmagnetic and 122 collinear spin point groups. Treating orbital and spin degrees of freedom independently, we identify the symmetry-allowed multipoles associated with nonmagnetic and collinear magnetic orderings and clarify their hierarchy through comparisons among the corresponding spin point groups. The resulting classification provides a symmetry-based database that distinguishes structural and magnetic contributions to active multipoles and directly identifies microscopic order parameters, including those responsible for $d$-, $g$-, and $i$-wave altermagnetism. Furthermore, by classifying response tensors within the same framework, we establish a correspondence between active multipoles and symmetry-allowed physical responses. This correspondence systematically distinguishes nonrelativistic responses that survive without spin-orbit coupling from those requiring relativistic effects, providing a unified framework for understanding and predicting spin-dependent electromagnetic and transport phenomena in magnetic materials.
Information Retrieval
AtomRec: Evolving Atomic Memory for Agentic Recommendation
Agentic recommender systems use large language models to maintain semantic memory and support evidence-aware recommendation. However, existing memory mechanisms often compress user and item information into coarse summaries and connect them with scalar collaborative links, making it difficult to preserve fine-grained preference stages or retrieve interpretable evidence as user interests evolve. We propose AtomRec, an agentic recommender with evolving atomic collaborative memory. AtomRec represents user and item memories as structured atomic units, builds semantic links across related memories, and evolves related historical fields when new interactions arrive. During recommendation, it retrieves linked memories as multi-hop evidence paths rather than isolated neighbor summaries, allowing collaborative signals to support grounded ranking. Experiments on four public benchmarks show that AtomRec consistently outperforms state-of-the-art agentic and memory-augmented baselines, with around 8.5% average relative improvement across metrics.
Read the full abstract
Machine Learning
From Deep to Shallow: Unconstrained and Efficient Layer Merging Strategy
Although Deep Neural Networks have become foundational in many areas of Machine Learning, high computational demands limit their application in resource-constrained environments. To address this issue, depth compression methods have been proposed to identify and linearize redundant activation functions, thereby allowing for the merging of layers without intermediate non-linearities. However, these methods face two key challenges: they cannot be directly applied to convolutions with padding due to the absence of an analytical solution for merging these layers, and they typically increase the kernel size of merged layers, thus limiting speed-up gains. To overcome these limitations, we propose an efficient strategy that enables merging of layers without an existing analytical solution, and also without increasing kernel size. We validate our approach across multiple architectures and datasets, and measure inference speed-up gains on real embedded platforms. We publicly released the code at https://github.com/ShulzhenkoPetr/deep-to-shallow.
Read the full abstract
Artificial Intelligence
Reinforcement Learning for Sequential Solar PV Policy Design under Uncertainty: An Agent-Based Approach
Designing effective and fiscally sustainable policies for solar photovoltaic (PV) adoption requires balancing adoption gains against public expenditure under uncertainty and heterogeneous decision-making. This study formulates PV policy design as a sequential decision problem and integrates reinforcement learning (RL) with a stochastic agent-based model (ABM) that simulates yearly solar PV adoption under uncertainty. A policymaker agent selects annual incentives, including capital grants, subsidised loan rates, and feed-in tariffs, over a 16-year horizon. Adoption–cost trade-offs are explored by varying policy preferences within a scalarised reward framework. Policies are learned using PPO, SAC, and TD3 and evaluated under stochastic simulation. The results show that this approach produces a clear trade-off structure: the highest-adoption policy (TD3, $w_{\text{cost}}=0.5$) achieves approximately 4,145 adopters at a cost of EUR 41.73 million, while the lowest-cost policy (PPO, $w_{\text{cost}}=2.0$) reduces expenditure to EUR 7.27 million with 2,682 adopters. The balanced policy (PPO, $w_{\text{cost}}=1.6$) achieves 3,495 adopters at a cost of EUR 22.47 million. Across algorithms, consistent trade-off patterns are observed, indicating robustness of the adoption–cost relationship. Compared with static baseline policies, the RL framework explores a broader range of policy configurations. These findings demonstrate the potential of RL as a flexible tool for adaptive policy design under uncertainty.
Read the full abstract
Information Theory
Near-Field Physical-Layer Authentication Under Impersonation Attacks
This paper studies physical-layer authentication (PLA) in the near-field regime under impersonation attacks. Unlike the far-field case, where the steering vector depends only on the angle of arrival (AoA), in the near field it depends on both angle and distance. We analyze the attack by minimizing the mean-square error (MSE) between the signal received from a legitimate transmitter Alice and the signal generated by an active attacker Eve. For a single-antenna Eve, we derive the optimal scalar precoder and show that, for an inter-element spacing no larger than half a wavelength and under the standard second-order Fresnel approximation, perfect impersonation is possible only if Eve has the same angle and the same distance from Bob as Alice. We then extend the analysis to a multi-antenna Eve and derive the optimal precoding vector. In this case, perfect impersonation is possible only if Alice steering vector belongs to the subspace spanned by Eve steering vectors. Simulation results show that, in the near field, a distance difference is sufficient to prevent a successful impersonation attack even when Alice and Eve have the same AoA, and confirm the analytical results.
Read the full abstract
Cryptography and Security
ReCAST: Restoration-aware Cascaded Stage-wise Training for Obfuscated SMS Risk Classification
Fraudulent messages sent via Short Message Service (SMS) are increasingly obfuscated to evade cost-conscious classifiers in production systems. In Chinese SMS, attackers can exploit a wide range of carefully crafted obfuscation strategies to hide risk-bearing phrases while preserving human readability, making direct classification brittle under real-world latency and throughput constraints. We propose ReCAST, a Restoration-aware Cascaded Stage-wise Training framework for robust obfuscated Chinese SMS classification. ReCAST distills a large teacher model's de-obfuscation ability into a smaller deployable student model by supervising obfuscated span detection, obfuscation type prediction, and text restoration, and then uses the restoration-aware student for downstream risk classification. Experiments on an internally constructed real-world Chinese SMS benchmark show that ReCAST substantially improves classification performance over directly trained baselines under obfuscation. The results suggest that restoration-aware distillation offers a practical path toward robust SMS risk classification with smaller deployable models under production-oriented constraints.
Read the full abstract
Artificial Intelligence
MARLA: A Conceptual Scaffold for Regulatory Learning under the EU AI Act
The EU AI Act positions regulation as part of the infrastructure for safe, trustworthy and market-ready innovation. Realising this ambition requires regulatory learning: the evidence generated during implementation must be translated into governance and legal knowledge that supports consistent interpretation, effective oversight, and adaptation as technologies evolve. Yet the actors who produce this evidence and those who rely on it operate in different professional worlds. This paper proposes MARLA (Map, Assess, Report, Learn, Adapt), a conceptual scaffold organising regulatory learning as a five-stage cycle centred on the implementation of legal requirements into socio-technical practices, situated at the Local, National and European levels of the AI Act's governance architecture. Deliberately non-prescriptive, MARLA gives technical and legal stakeholders a shared vocabulary in which each of the first three stages generates its own documentable form of regulatory learning. We illustrate the scaffold with two piloted case studies and a prospective National-to-European illustration.
Read the full abstract
Signal Processing
AFDM-Enabled ISAC in Dynamic Environments: Fundamentals, Technologies and Opportunities
Dynamic environments pose fundamental challenges to integrated sensing and communication (ISAC), particularly due to severe Doppler effects, rapidly time-varying channels, and the intricate coupling between delay and Doppler shifts. Affine frequency-division multiplexing (AFDM), with its inherent capability of characterizing and separating delay and Doppler effects, has emerged as a promising waveform for dynamic ISAC. This article provides a comprehensive overview on AFDM-enabled ISAC in dynamic environments, covering its fundamental principles, distinctive advantages, representative application scenarios, and key enabling technologies. We first characterize the key features of ISAC in dynamic environments and introduce the fundamentals of AFDM, followed by an analysis of scenarios where AFDM can provide significant performance benefits. Then, several key enabling technologies for AFDM-based ISAC in dynamic environments are elaborated upon, accompanied by case studies on the critical aspects therein. Finally, open challenges and promising future research directions are discussed, aiming to provide a comprehensive reference for researchers and practitioners while inspiring further innovation in this emerging field.
Read the full abstract
Cryptography and Security
Forgetting Without Restarting: Execution-State Unlearning for Stateful LLM Agents
Long-running LLM agents are stateful: beyond the transcript they accrete compressed summaries, plaintext memory, pending tool plans, and, under every serving API, a KV cache. Yet today's "forget" operations delete a plaintext memory record and stop, leaving every artifact derived from the revoked information intact. We formalize execution-state unlearning: after a forget request, the agent must behave as if it had never observed the target. Modeling the runtime as a deterministic transition system, we prove that the pre-target trajectory prefix is shared with this counterfactual world for free, that the post-target suffix is irreducibly tainted without token-level attribution, and that exact unlearning requires at least $T-\tau+1$ recomputed transitions, where $\tau$ is the target's injection step. Provenance-Guided Selective Replay attains this bound as a cross-layer contract spanning prompt, compressed memory, and cache: a provenance graph locates the injection point, checkpoint restoration reduces to cropping the KV cache, and sanitized replay regenerates the counterfactual suffix. Audited with elicitation, stochastic, and string-free behavioral tests across three agent suites, nine baselines, and three model families, memory deletion leaves leakage unchanged, instruction-based forgetting collapses under elicitation (Leak@probes = 1.00), and source redaction still acts on a revoked preference in 80% of episodes, while selective replay is indistinguishable from a full reset at up to 9x fewer recomputed tokens.
Read the full abstract
Networking
A Wavelength Borrowing Architecture for Optical Data Center Networks - Extended Version
The growth of east-west traffic, along with the cost and power consumption of electronic switching, is motivating the integration of a low-power, high-rate, all-optical layer within the data center network. This paper presents a spine-leaf all-optical architecture in which the default wavelength configuration, one wavelength per source–destination leaf pair, can be reconfigured to accommodate unbalanced traffic demand: wavelengths that are unused or lightly loaded at one leaf are borrowed by another leaf with higher demand. This topology-engineering capability is combined with a traffic-engineering scheme, based on two-hop detouring, enabling the control of wavelength load while limiting the amount of detoured traffic. A key feature of the architecture is that its degree of wavelength reconfigurability is set by a single tunable parameter, the borrowing degree $B$, ranging from none to full; performance evaluation shows that performance nearly equal to that of a fully reconfigurable architecture is achieved well below the maximum $B$, saving the complexity of a fully reconfigurable solutions. The optical fabric relies on mature, data-center-grade components, namely AWGs, AWGRs, colorless OXCs, and combiners. The reconfiguration speed and the adopted dynamic control policy make the architecture suitable for network tiers where the traffic varies on timescales of tens of minutes or more, e.g., among groups of racks (pods).
Read the full abstract
Pattern Formation and Solitons
Evolution of instability fronts in sine-Gordon equation dynamics
Solutions of the Whitham modulation equations for one-phase periodic waves obeying the sine-Gordon equation are found that describe the evolution of an oscillatory region behind an instability front propagating into the instability region. A simple self-similar solution describes the whole region between two fronts of instability resulting from a localized initial disturbance in the unstable state. Another hodograph solution represents typical waves close to the instability fronts. This theory generalizes the approach used previously for systems obeying the nonlinear Schroedinger equation.
Read the full abstract
Digital Libraries
The Generative AI Gold Rush in Theoretical and Computational Research
Generative AI is changing the production conditions of theoretical and computational research, but its sys tem level effects require measures that separate plat form growth, field specific divergence, and production structure. We assemble 2,080 monthly observations for twenty arXiv archives from January 2018 through Au gust 2026 and a separate pseudonymized Mathematics author panel. A regularized convex synthetic control fitted through December 2025 identifies the January August 2026 anomaly, while spatial placebos, prior year pseudo holdouts, donor refits, and alternative preperiods assess comparative robustness. Mathematics recorded 47,127 list entries, 33.5% above 2025 and 11.9% above a synthetic counterfactual of 42,113 entries. Qualified donor and preperiod designs yield 9.6% to 14.9%, and Mathematics has the largest RMSPE ratio among fifteen eligible placebo archives. Subfield growth is broad, with 29 of 30 primary math.* categories expanding. The author panel shows a marked thickening of the repeated output tail. The share of active author units produc ing at least five submissions rose from 2.45% to 3.80%, while the ten submission tail rose from 0.21% to 0.49%. These results document a new and unusually large 2026 Mathematics production regime shift. Its timing and production structure, combined with independent evi dence on AI diffusion and verifiable research tasks, are consistent with delayed diffusion and capability thresh old mechanisms. The comparative design identifies the anomaly, and separate triangulation evaluates AI related explanations. The findings locate verification, selection, and attention as central constraints for research gover nance.
Read the full abstract
Artificial Intelligence
AutoLR: Automating the Path from Research to Launch Review in Industrial Recommender Systems
Improving an industrial recommender is an iterative research-and-engineering process rather than a direct path from idea to deployment. In DASHEN, NetEase's gaming-community app, algorithm engineers typically identify promising directions from research papers, technical reports, and prior production experiments; reproduce or adapt the underlying methods; implement them in the production codebase; and evaluate the resulting models through training and offline experiments. Promising candidates are then advanced to online A/B tests, and those demonstrating robust gains are submitted to Launch Review—the internal gate for full-traffic rollout. Large language models (LLMs) can assist with individual stages of this workflow, but the overall process remains human-dependent without a harness that can reliably coordinate them across long-running, often multi-day experimental cycles. We present AutoLR, initially built as Auto Launch Review and later extended upstream into an autonomous research-to-launch harness. AutoLR combines three system mechanisms: a multi-expert council that debates and adversarially reviews proposals; a deterministic evidence-weighted exploration–exploitation selector that allocates a limited trial budget across candidate directions and uses Council reranking; and a layered knowledge system that combines external research, production-system knowledge, and DASHEN-specific domain knowledge—such as game communities, player characteristics, and content-interaction patterns—with posterior evidence from configurations, patches, logs, failures, and offline outcomes. LLM agents perform semantic reasoning and code generation, while deterministic controllers retain authority over execution, metric extraction, guardrails, and persistent state transitions.
Read the full abstract
Artificial Intelligence
CHAMP: Cross-domain Hybrid Architecture for Matchmaking and Prediction in Online Multi-Player Games
Multiplayer Online Battle Arena (MOBA) games rely on matchmaking to maintain competitive balance. Our prior work, CUPID, framed matchmaking as an assignment re-optimization problem and showed that a single-mode win-rate predictor can meaningfully rebalance teams. However, deploying such a system across diverse player populations exposes three practical bottlenecks: most queueing players lack sufficient in-mode match history (cold start), skill distributions shift drastically across rank tiers (distribution inconsistency), and extreme skill segments are severely data-starved. We present CHAMP, a cross-domain matchmaking framework that resolves these deployment bottlenecks. To address data sparsity and cold starts, CHAMP replaces the target-mode-only player profile with a hybrid domain feature collection: a timestamp-ordered cross-mode short-term sequence whose slices are annotated with target-domain features, plus per-mode breakdowns of long-term, real-time and team statistics. We further propose the Domain-Aware Win-rate Network (DAWN): a Domain-aware Knowledge Extractor (DAKE) compiles target-mode attributes into learnable representations that feed Domain-Aware Temporal/Spatial/Permutation OmniNet Encoders (DATOE/DASOE/DAPOE), so that mode-conditioned representations and per-mode debiasing are learned jointly inside a single shared network. Online, one trained DAWN serves every supported mode, with per-mode position-satisfaction thresholds as the only mode-specific knob. Offline, DAWN achieves 67.73% win-rate prediction accuracy, outperforming all evaluated attention and sequence baselines. Online A/B tests across the entire League ladder of a large-scale MOBA game, from novice players up to the top-expert players served by Elite Mode, demonstrate consistent drops in imbalanced matches. For lower-tier players, CHAMP reduces the 5-minute kill crushing rate by up to 20.73%.
Read the full abstract
Artificial Intelligence
From Interaction Traces to Persistent Skills: Online Evolution for Computer-Use Agents
Computer-use agents can execute increasingly complex tasks in graphical interfaces, but their interaction experience is typically transient: procedural knowledge acquired from one rollout is not systematically retained, refined, and reused in later tasks. Existing skill libraries provide external procedural knowledge, yet their incremental value over the same agent operating without skills, as well as their longitudinal dynamics under repeated interaction, remain insufficiently characterized. We present an online skill-evolution framework that converts interaction trajectories and evaluator feedback into a persistent, versioned library of reusable procedures. Each iteration executes against a frozen library snapshot, and evidence-guided skill updates become available in subsequent iterations without changing model parameters. We compare the full evolving-library system with a configuration-matched empty-library control across four OSWorld application domains under the same fixed action-generation and GUI-grounding stack, task sets, and iteration horizons. Following a five-iteration empty-library warm-up, Full attains a higher post-warm-up mean evaluator score in all four observed domain runs, with mean differences ranging from 5.7 to 18.6 percentage points and domain-dependent temporal stability. In GIMP, provenance-aware analysis reveals retrieval across task-of-origin boundaries and revision churn, where repeated accepted edits fail to recover the originating task. These findings characterize evolving skill libraries as auditable, shared procedural memory that can improve a fixed computer-use stack, while showing that their benefits are conditional and repeated revision does not guarantee recovery. Code is released at https://github.com/LongtaoHu/Skill-Evo4GUI.
Read the full abstract
High Energy Physics (Phenomenology)
Higgs Boson Pair Production via Gluon Fusion: Higher-Order Corrections and Theoretical Uncertainties
In this contribution, the higher-order QCD and electroweak corrections to Standard Model Higgs boson pair production via the gluon-fusion mechanism, $gg\to hh$, are summarized and the different sources of theoretical uncertainty are assessed. The discussion includes finite top quark mass effects, matching to parton showers, approximate NNLO and N$^3$LO QCD corrections, NLO electroweak effects, and uncertainties associated with the top quark mass scheme and perturbative scale choices. In addition, we provide an updated state-of-the-art recommendation for the inclusive gluon-fusion Higgs boson pair production cross section and the corresponding Higgs boson pair invariant-mass distribution.
Read the full abstract
Multimedia
PRISM-Bench: An Audio-Centric Diagnostic Benchmark for Text-to-Audio-Video Generation
Text-to-audio-video (T2AV) generation has advanced rapidly, but its evaluation still underestimates the audio modality. Existing benchmarks either treat audio as an auxiliary component of video quality or assess it in isolation from audiovisual grounding, making it difficult to diagnose where current systems truly succeed or fail in audio generation. We present PRISM-Bench, the first audio-centric diagnostic benchmark for T2AV generation. Built from a rigorously curated dataset of 900 human-verified samples, PRISM-Bench factorizes audio evaluation along two orthogonal axes: audio type (Speech, Music, and Sound) and sound-source visibility (On-screen vs. Off-screen). It evaluates generated content across four perceptual dimensions (Audio-Visual Coherence, Audio Quality, Audio Expressiveness, and Prompt Following) with 35 fine-grained criteria. To ensure reliable assessment, we adopt an enhanced MLLM-as-a-Judge protocol based on blind, side-by-side comparison against ground-truth references, demonstrating strong alignment (over 70% mean agreement) with human raters. Our evaluation of recent T2AV systems highlights a significant performance gap between frontier and open-source models. Furthermore, we demonstrate that current generation paradigms overfit to perceptual fidelity while struggling with complex grounding and control tasks, particularly in generating music and synchronized On-screen audio.
Read the full abstract
Artificial Intelligence
LLM-Assisted Behavioural and Scenario Augmentation for Agent-Based Energy Adoption Models
Recent advances in large language models (LLMs) create opportunities to enrich simulation-based energy policy analysis, particularly by supporting structured behavioural assumptions and exploratory techno-economic scenarios. However, directly replacing adoption models with LLM reasoning raises concerns regarding interpretability, reproducibility, and behavioural validity. This paper proposes a hybrid framework for LLM-assisted specification design, integrating bounded behavioural rubrics and structured scenario specifications into a calibrated agent-based model (ABM) of solar photovoltaic (PV) adoption by Irish dairy farms. The proposed approach preserves the original techno-economic adoption mechanism while augmenting it with bounded behavioural modulation and scenario-driven uncertainty analysis. Behavioural effects are represented through interpretable conservative, balanced, and optimistic rubrics, while future policy and market conditions are explored through fixed, rule-validated scenario specifications. Experimental results across multiple policy settings, Monte Carlo worlds, and random seeds demonstrate stable and economically plausible behaviour, with adoption outcomes remaining bounded and monotonic across behavioural regimes. The framework achieves up to approximately 13% behavioural adoption increase relative to the corresponding logistic case without producing unstable or unrealistic saturation dynamics. The results demonstrate that LLM-assisted specifications can be integrated into calibrated energy ABMs in a controlled, reproducible, and policy-relevant manner.
Read the full abstract
Artificial Intelligence
CoSkill: Joint Reinforcement Learning of Reasoning and Meta-Skill Agents for Hierarchical Skill Evolution
Skill libraries improve the sample efficiency of agentic reinforcement learning (RL) by enabling large language model (LLM) agents to reuse procedural knowledge. Yet existing paradigms exhibit structural shortcomings: they either decouple skill evolution from policy optimization or instantiate meta-skills as fixed workflows. Both treat skills as passive objects to be managed, limiting the flexible evolution of skills and their co-adaptation with the reasoning agent. To address the limitations, we propose CoSkill, a unified multi-agent RL framework that recasts the static meta-skill workflow as a learnable Meta-Skill Agent and jointly trains it with a Reasoning Agent over a hierarchical skill library. By modeling the Reasoning and Meta-Skill Agents as a cooperative team sharing a single backbone, CoSkill enables end-to-end co-adaptation: the Reasoning Agent conditions its actions on a retrieved task skill and step skills selected from its child set, while its task performance guides the Meta-Skill Agent in refining those step skills. Experiments on ALFWorld and WebShop show that CoSkill substantially outperforms prior skill-based and RL baselines, achieving success rates of 98.4% and 90.6%, respectively (+3.5 and +6.2 pp). As shown in Figure 1, CoSkill achieves superior early-stage sample efficiency, asymptotic performance, and wall-clock efficiency. Our code is available at https://github.com/jinyuan-cookie/CoSkill.
Read the full abstract
Artificial Intelligence
MZ-Rain: Moisture-Budget-Guided Zero-Inflated Model for Station-Level Precipitation Nowcasting
Accurate station-level precipitation nowcasting is critical for agriculture, water resource management, and disaster prevention, which typically is formulated as a time series forecasting problem. However, conventional time-series modeling techniques face two major challenges in addressing station-level precipitation nowcasting: (1) Lack of Physics-Guided Modeling}, where meteorological variables are treated as a homogeneous set without accounting for their distinct roles in precipitation formation, leads to predictions that deviate from the physical processes governing precipitation. (2) Severe zero inflation in precipitation, where dry intervals dominate the dataset, obscuring meaningful precipitation patterns and complicating the predictive modeling. To address these challenges, we propose MZ-Rain, a moisture-budget-guided zero-inflated sLSTM framework for station-level precipitation nowcasting. Guided by the moisture budget equation, MZ-Rain decomposes the precipitation formation process into process-specific pathways corresponding to moisture storage, moisture transport, surface evaporation, and precipitation persistence, and captures their temporal evolution through dedicated sLSTM branches. To account for the zero-inflated nature of precipitation, MZ-Rain introduces an adaptive Tweedie modeling strategy that adaptively modulates the rainfall mean while jointly learning precipitation occurrence as an auxiliary task, enabling the model to better balance dry-wet discrimination and quantitative precipitation estimation. Extensive experiments across diverse geographical and climatic regimes demonstrate that MZ-Rain consistently outperforms strong baselines on multiple evaluation metrics, including CSI, FAR, MSE, and MAE. In particular, the model exhibits superior skill in forecasting heavy precipitation events, while benefiting from physically grounded process modeling.
Read the full abstract
Optics
Frugal Effective Models for Nanophotonic Scattering: Optimizing Global Polarizability Matrices for Metasurface Design
Accurate nano-photonics simulations of large scale devices like optical metasurfaces require high accuracy reduced models for the device constituents. We present an automated framework for the optimization of Global Polarizability Matrix (GPM) models, which represent a complex scatterer as a small set of non-local effective dipoles. Our goal is to find the most frugal model that reproduces a particle's scattering response within a user-defined accuracy. The method iteratively removes redundant dipoles while re-adapting the positions of the remaining ones via gradient based optimization, stopping at the smallest model that still meets the target. Automatic differentiation, combined with an untrained neural network that reparametrizes the dipole positions, helps to place the dipoles at physically intuitive locations. We demonstrate the versatility of this approach across diverse geometries, from two dimensional ridges over simple spheres to complex three-dimensional particles, achieving compression factors of typically two orders of magnitude compared to full-wave simulations, for target accuracies in the order of few percent. We finally demonstrate how accurate, frugal effective models enable large-scale meta-deflector optimization without periodic approximations. This robust recipe for constructing frugal effective models paves the way for the rapid simulation of large-scale photonic assemblies, required for example for metasurface design.
Read the full abstract
Information Retrieval
Personalized Task Dependency Graphs for Mitigating Signal Erosion in Multi-Task Recommendation
Optimizing multiple conversion objectives is a core challenge in industrial recommendation, often limited by signal erosion in rigid architectures. Existing Multi-Task Learning (MTL) methods typically enforce uniform dependency strengths across a static conversion funnel, overlooking how task correlations naturally vary based on item characteristics. Hierarchical message passing along these fixed chains leads to cumulative signal attenuation, which degrades performance on sparse, deep-funnel objectives. To address this, we propose the Personalized Task Dependency Graphs (PTDG). While respecting necessary physical causal constraints (e.g., Click -> Pay), PTDG dynamically "rewires" the intensity of dependency pathways for each item via low-rank approximation to ensure structural robustness. We implement a GCN-based propagation with hard causal masking to establish adaptive information shortcuts. Additionally, we introduce an Adaptive Progressive Masking (APM) strategy that decouples shared parameters according to task sparsity, helping to stabilize optimization. Experiments on KuaiRand1K and an industrial dataset show that PTDG significantly improves AUC on sparse conversion tasks by up to 1.45%, while maintaining comparable performance on dense objectives. Online A/B testing shows PTDG improves Conversion Rate (CVR) by 1.2% and effective Cost Per Mille (eCPM) by 1.9% relative to the baseline.
Read the full abstract
Machine Learning
When Genomic Masking Priors Fail to Transfer: Strong Variant Prediction, Weak Functional Generation
Bidirectional discrete diffusion model appears naturally suited to genomic modeling because it can reconstruct missing sequence from both flanks. We developed GenDA (Genomic Density-optimized Absorbing Diffusion) under the additional hypothesis that entropy-guided span placement would concentrate reconstruction pressure on compositionally complex regions, improving both downstream variant-effect prediction and functional sequence generation. Our results only partially support this premise. After supervised fine-tuning, the 202M-parameter GenDA model reaches a pooled ClinVar SNV AUROC of 0.774, exceeding a similarly scaled autoregressive model by 0.103. However, a matched random-span variant reaches 0.777, providing no evidence that entropy guidance causes the ClinVar improvement. More unexpectedly, GenDA fails a zero-shot functional inpainting stress test: across promoters, enhancers, exon boundaries, and intron boundaries, it does not consistently outperform a control that shuffles the native gap while exactly preserving 3-mer composition. Failure is already present for 50–500-bp gaps, although enhancer degradation worsens at longer gaps. Diagnostics identify several boundary conditions: entropy measures local sequence complexity rather than functional importance; 1-mer tokenization limits physical context; training spans are capped at 300 bp; and high absolute AlphaGenome fidelity can coexist with negative control-normalized restoration. These results show that strong fine-tuned variant prediction, a plausible corruption prior, and functional generation are distinct claims that require separate validation.
Read the full abstract
Computer Vision
Mitigating Performance Discrepancy in Cross-Domain 3D Class-Incremental Learning
3D perception plays a crucial role in real-world applications such as autonomous driving, robotics, and AR/VR. In practical scenarios, 3D perception models need to continually adapt to newly emerging 3D object categories, making class-incremental learning (CIL) particularly important. However, unlike 2D images, 3D point clouds are inherently heterogeneous: objects from the same class may not only come from the clean CAD domain, but also from RGB-D camera scans of varying quality, video reconstructions, or even corrupted observations. We discover that such heterogeneity introduces a new challenge beyond catastrophic forgetting: the degree of performance degradation can vary substantially across domains, a phenomenon we term performance discrepancy. To investigate this problem, we establish the Domain3D-CIL training and evaluation protocol, which contains point cloud categories from heterogeneous domains. We further adapt a wide range of mainstream CIL methods to the 3D modality. The results demonstrate that this performance discrepancy consistently appears across these baselines. To mitigate this issue, we introduce PolyMem, an exemplar-free approach that implicitly models rich high-order statistics of the feature distribution to enhance cross-domain robustness. Experiments demonstrate that our method effectively alleviates the performance discrepancy while improving the model's performance across domains. Code will be made publicly available upon acceptance.
Read the full abstract
Artificial Intelligence
MM-IFEval-Pro: A Multilingual and Attack-Resistant Benchmark for Instruction-Following in Vision-Language Models
As vision-language models (VLMs) rapidly advance in image understanding, cross-modal reasoning, and complex instruction execution, instruction-following capability has become a key indicator of their reliability and practicality. However, existing multimodal instruction-following benchmarks still suffer from limited language coverage and insufficient adversarial safety scenarios, making them inadequate for evaluating real-world multilingual and safety-sensitive settings. To address these gaps, we present MM-IFEval-Pro, a multimodal instruction-following benchmark covering Chinese and English tasks as well as diverse instruction hijacking cases. MM-IFEval-Pro includes 4 major task categories and 24 subcategories and 8 instruction categories with 52 subcategories, with each sample containing an average of 3.0 constraints to realistically simulate complex instruction scenarios. We further construct a reinforcement-learning training set enriched with Chinese and adversarial instructions, which significantly improves model performance on MM-IFEval-Pro and transfers effectively to other mainstream multimodal benchmarks, demonstrating strong cross-task and cross-language generalization.
Read the full abstract
Combinatorics
Characterizing pyramidal Hadamard designs with the largest number of fixed points
A symmetric $(v,k,\lambda)$-design is said to be $f$-pyramidal, with $f<v-1$, under the action of a group $G$ if $G$ acts as an automorphism group fixing $f$ points and acting sharply transitively on the remaining ones. We show that, necessarily, $f\leq v -2(k-\lambda)$. In particular, for a Hadamard design with parameters $(v,k,\lambda)=(4m-1, 2m-1, m-1)$, for some $m\in\mathbb{N}$, it follows that $f\leq 2m-1$. Recently, working on the complement design, the family $P$ of Hadamard $(4m-1,2m-1,m-1)$-designs admitting an abelian $(2m-1)$-pyramidal automorphism group $G$ has been completely determined in the case $m=2^{k}>1$. In this paper, we generalize that result by showing that $P$ coincides with the family of Hadamard $(4m-1,2m-1,m-1)$-designs admitting a $(2m-1)$-pyramidal automorphism group, without assuming either that the group is abelian or that $m$ is a power of $2$.
Read the full abstract
Astrophysics of Galaxies
Membership Determination of 45 Open Clusters Beyond 3 kpc
Context. Open clusters (OCs) are fundamental tracers for studying the structure and chemical evolution of the Galactic disc. Reliable identification of cluster members, particularly at the faint end of the main sequence, remains challenging owing to severe field-star contamination and the declining photometric completeness for distant, obscured systems. Aims. We aim to identify and characterise distant OC candidates at heliocentric distances $d \gtrsim 3$ kpc using Gaia DR3 astrometry, with emphasis on improving the recovery of faint candidate members in highly contaminated stellar fields. Methods. We apply a probabilistic membership method based on Gaia DR3 proper motions and parallaxes. Candidate members are selected through combined astrometric criteria and refined via spatial filtering using radial density profiles, with photometric data serving as a consistency check. For clusters affected by significant extinction, near-infrared photometry from 2MASS is incorporated to compensate for the limitations of optical data. Results. We analyse a sample of 45 targets, comprising 30 previously reported OCs and 15 newly identified candidates, all located at $d \gtrsim 3$ kpc. The method yields more compact astrometric distributions and improves the recovery of faint candidate members in several systems. However, residual field contamination remains non-negligible for the most obscured clusters projected against the Galactic plane. Conclusions. Gaia DR3 astrometry provides an effective basis for the exploratory identification and first-order characterisation of distant OC candidates. Dedicated near-infrared follow-up observations will be essential for robust membership assessment in high extincted regions.
Read the full abstract
Soft Condensed Matter
The elasticity of semiflexible polymers with reversible spontaneous curvature in two dimensions
Many semiflexible filaments exhibit structural asymmetries that lead to a curved ground state (curvature in the absence of thermal fluctuations, external forces, or torques). This property is known as spontaneous curvature. Many semiflexible polymers can reversibly develop spontaneous curvature through internal conformational transitions or through transient interactions with their molecular environment. In this article, we show that such reversible spontaneous curvature, even in the absence of transitions in the value of the bending stiffness, provides a minimal mechanism for ensemble-dependent elasticity. We formulate an analytically tractable two-state modified wormlike-chain model in two dimensions, in which an uncurved state (without spontaneous curvature) competes with a curved state (possessing spontaneous curvature). The transition is controlled by an activation energy. Within the weak-bending approximation, we obtain the Gibbs and Helmholtz partition functions for a stretched filament with reversible constant (uniform along the backbone) or sinusoidal spontaneous curvature and for a grafted bistable filament driven either by an end torque or a bending force. Stretching or bending induces a crossover from spontaneous-curvature-dominated to entropic elasticity. We show how the elastic response differs in the two ensembles (Gibbs vs Helmholtz). In addition, we consider the case where the curved state is also characterized by a higher value of the bending stiffness. In that case, a reentrant transition is possible as we stretch the bistable filament.
Read the full abstract
Computation and Language
CC-Mediation: Evaluating Large Language Models for Cross-Cultural Conflict Mediation
Cross-cultural mediation by large language models (LLMs) requires deciding both when to intervene and how to respond in culturally grounded conflicts. Progress on this problem has been limited by the lack of (1) mediation datasets with measurable downstream effects and (2) principled metrics for evaluating intercultural stance change. To address these gaps, we introduce CC-Mediation, a cross-cultural mediation benchmark of $1{,}661$ ten-turn dialogues grounded in the Developmental Model of Intercultural Sensitivity (DMIS), containing culturally grounded conflicts, mediation interventions, and post-intervention trajectories. We further propose two DMIS-based evaluation metrics: Trajectory AUC, which measures the persistence of intercultural improvement over time, and a signed Wasserstein-1 distance, which measures the magnitude and direction of shifts in intercultural stance. Both metrics show strong agreement with human judgment of DMIS-grounded stance shift. Using CC-Mediation, we find that current LLMs have limitations on both axes: intervention timing (when) failure stems from a positional prior that ignores dialogue content, while mediation strategy (how) failure arises from a late-layer elicitation collapse rather than a knowledge deficit.
Read the full abstract
Mesoscale and Nanoscale Physics
Excitation caging in a vertex-frustrated quasiperiodic Einstein artificial spin ice
Naturally occurring bulk quasicrystals are rare, and magnetic instances are rarer still, with chemical constraints typically permitting the synthesis of approximants rather than true quasicrystalline magnets. Here, we present an artificial spin ice based on a recently discovered Einstein lattice, the Hat tiling, which is built from the first known shape - the hat - that tiles the plane only aperiodically. The Einstein artificial spin ice has long-range structural order with no translational symmetry, and low-connectivity vertices with well-defined local ground states and excitations. Together, these properties provide a model two-dimensional quasicrystalline magnet, with magnetic correlations that we probe with magnetic force microscopy and parallel-tempered Monte Carlo simulations. We identify a two-stage partial ordering process, driven by the competition between two possible positions for magnetic excitations. This competition is resolved in the ground-state manifold, where exactly one magnetic excitation is caged on each antihat, yet the manifold remains macroscopically degenerate. This yields an unusual type of medium-range order, where the underlying quasiperiodic long-range order is randomly modulated by a strictly constrained disorder. Our findings establish the Einstein artificial spin ice as a blueprint for understanding quasicrystalline magnetism, demonstrating how quasiperiodic monotile geometries can be exploited to engineer unconventional magnetic phases with no direct equivalent in periodic systems.