arXiv:2605.25941v1 Announce Type: new
Abstract: Text-to-video diffusion transformers encode semantic information unevenly across model depth, which constrains effective concept erasure. We identify a representational bottleneck, termed concept-layer topological alignment, under which target concepts exhibit higher separability at certain representational depths. Outside these depths, concept and non-target signals remain strongly entangled, limiting the effectiveness of depth-specific erasure. This observation reframes concept erasure as the problem of identifying representational depths where concept-non-target separation naturally emerges. Motivated by this structural constraint, we introduce CLEAR, a separability-driven optimization framework for concept erasure that explicitly enforces concept-layer alignment. CLEAR operationalizes this principle by formulating layer selection as an optimization problem over concept-non-target separability, rather than relying on layer-agnostic or heuristic choices. To enable this, we introduce a separability-aware objective that favors layers exhibiting stronger concept-non-target separation. Experiments on large-scale text-to-video diffusion models demonstrate that enforcing concept--layer alignment leads to more precise concept suppression while preserving overall generative quality.
Science Journals
arXiv:2605.24899v1 Announce Type: new
Abstract: Ontologies represent the conceptual knowledge of a domain. At the core of an ontology is the taxonomy of concepts and subconcepts that represent specific entities, which can be complex to build. In many cases, information is available in the form of records describing the characteristics of relevant entities, i.e., tabular data. Identifying patterns and similarities in such data can serve as a basis for identifying concepts and organizing them. However, doing so manually can be challenging, and purely automatic approaches, such as agglomerative clustering or relying on a large language model to analyze the data, can leave the user with overwhelming results and little control. In this paper, we describe a tool that enables the progressive and interactive construction of a taxonomy of concepts by identifying clusters as well as their intentional definitions. To do so, we rely on weighted self-organizing maps as a clustering method because they enable the creation of an arbitrary number of clusters that are distinct with respect to the distributions of values of specific characteristics of the clustered entities. We show that, by integrating this mechanism and others for rapidly creating concepts that group together instances from tabular data, this tool represents a middle ground between purely manual analysis and automatic methods for building ontological taxonomies.
arXiv:2605.25151v1 Announce Type: new
Abstract: Large language models are increasingly used as behavioral simulators, but it remains unclear when their outputs reflect human-like cognitive mechanisms rather than prompt-sensitive surface patterns. We study this question through the realization effect, a well-characterized finding in behavioral economics in which risk-taking differs systematically after paper versus realized gains and losses. We evaluate LLM behavior at three levels: prompt-only behavioral sensitivity, linear readout of internal representations, and causal control via activation steering. Prompt-only results show systematic condition sensitivity, but the directional pattern does not reproduce human realization-effect predictions. Gemma's residual stream contains a linearly decodable realization-status signal at layer 18 that generalizes to held-out prompts. Steering along this direction does not, however, reliably shift downstream risk choices, a null result that holds across positive scales and in a negative sign-symmetry run. Behavioral sensitivity, latent readout, and causal control are three distinct properties that do not automatically co-occur, and successful latent readout is insufficient evidence that a model behaviorally relies on a representation during downstream decision-making.
arXiv:2602.06717v2 Announce Type: replace
Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) is commonly based on group sampling to estimate advantages and stabilize policy updates. In practice, computational limits often rule out very large groups, so training proceeds with finite rollout sets that can reinforce only the correct behavior they expose. At practical group sizes, updates can miss rare-correct trajectories while still containing mixed rewards, concentrating probability on more common sampled solutions. We derive the probability of such prompt-local tail-miss events as a function of group size, showing non-monotonic behavior, and in the categorical abstraction characterize how unsampled-correct mass can shrink even as total correct mass grows. Motivated by this analysis, we propose a difficulty-aware scaling coefficient, inspired by Focal loss, that down-weights updates on high-success sampled groups. Empirically, categorical simulation illustrates the same effect in the categorical setting, Maze provides a single-solution test, and LLM experiments include a representative GRPO group-size sweep together with fixed-$N$ transfer across GRPO, DAPO, and CISPO. On Qwen2.5-7B at $N{=}8$, our method improves average math pass@256 from 64.1 $\rightarrow$ 70.3 (GRPO), 69.3 $\rightarrow$ 72.5 (DAPO), and 73.2 $\rightarrow$ 76.8 (CISPO); OOD pass@256 also improves in all three cases, without increasing group size or computational cost.
arXiv:2605.25119v1 Announce Type: new
Abstract: Domain adaptation aims to mitigate performance degradation caused by distribution shifts between a labeled source domain and an unlabeled or sparsely labeled target domain. Most existing approaches estimate domain discrepancy either in feature space or in prediction space. However, these single-perspective strategies overlook a critical problem under domain shift: the reliability of the signals used for alignment. In practice, both learned representations and semantic predictions may become unreliable, and treating all target samples equally can lead to misleading alignment and suboptimal transfer. We introduce trust-aware domain adaptation, a principled framework that models domain discrepancy through the reliability of feature and prediction signals. Central to our approach is the Joint Feature-Prediction Discrepancy (JFPD), a unified formulation that jointly captures representation divergence and prediction divergence while weighting their contributions by sample-specific trust. Trust is quantified via two complementary mechanisms: uncertainty-aware trust, derived from prediction entropy to suppress unreliable predictions, and semantic-alignment trust, computed from prototype similarity in feature space to emphasize well-aligned representations. By prioritizing confident and semantically consistent samples while down-weighting noisy or ambiguous ones, JFPD provides a reliability-aware estimate of domain discrepancy. We further integrate JFPD into a training objective that guides adaptation toward trustworthy regions of the target domain. Experiments on standard benchmarks demonstrate that the proposed framework consistently achieves superior adaptation performance and yields discrepancy estimates that correlate with target-domain error. This work addresses, for the first time, the importance of modeling trust in the interaction between features and predictions for domain adaptation.
arXiv:2605.24895v1 Announce Type: new
Abstract: NMR is uniquely endowed to analyze dynamics, with line shape and relaxation measurements covering timescales over several orders of magnitude. Further insight arises from pulse sequences like chemical exchange saturation transfer or relaxation dispersion, which facilitate, respectively, the detectability and shift characterization of even lowly populated states, and the pinpointing of the exact exchange rates. The present study demonstrates that Steady State Free Precession (SSFP) experiments involving a train of pulses with flip angle {\alpha} spaced by repetition times TR, combine valuable features from both these experiments. Indeed, in the presence of chemical exchanges, SSFP yields via its offset-dependent excitation and saturation profiles, detailed information about the number, the chemical shifts and the populations of the exchanging sites even when these involve multiple intermediates with dissimilar abundances. Simultaneously SSFP can provide, via its TR dependence, a controllable timescale yielding kinetic information over a variety of slow/intermediate/fast exchange rates. All this is theoretically demonstrated with the aid of a Liouville-space formalism examining the steady state in the presence of chemical exchange. This formalism leads in both the slow and fast exchange regimes to analytical predictions that match well brute-force numerical calculations, which lend themselves to rapid and accurate fittings of exchange rates, chemical shifts, and site populations. The basic features associated to this novel approach to examine chemical kinetics are experimentally verified on simple model compounds; potential extensions are briefly discussed.
arXiv:2510.02327v2 Announce Type: replace
Abstract: Real-time speech-to-speech (S2S) models excel at generating natural, low-latency conversational responses but often lack deep knowledge and semantic understanding. Conversely, cascaded systems combining automatic speech recognition, a text-based Large Language Model (LLM), and text-to-speech synthesis offer superior knowledge representation at the cost of high latency, which disrupts the flow of natural interaction. This paper introduces a novel hybrid architecture that bridges the gap between these two paradigms. Our framework processes user speech through an S2S transformer for immediate responsiveness while concurrently relaying the query to a powerful back-end LLM. The LLM's text-based response is then injected in real time to guide the S2S model's speech generation, effectively infusing its output with rich knowledge without the full latency penalty of a cascaded system. We evaluated our method using a speech-synthesized variant of the MT-Bench benchmark that consists of multi-turn question-answering sessions. The results demonstrate that our system substantially outperforms a baseline S2S model in response correctness, approaching that of a cascaded system, while maintaining a latency on par with the baseline.
arXiv:2605.25310v1 Announce Type: new
Abstract: Tool-using LLM agents produce trajectories whose calls form a directed dependency graph: earlier tool outputs supply arguments to later calls. Whether this execution structure is represented inside the model is unknown; prior structural probes have targeted static code or chain-of-thought text, not an agent's run-time call graph. A low-capacity edge probe on the residual stream of Qwen3-32B decodes the tool-call dependency graph well above both a Hewitt--Liang random-label control and a positional baseline. A counterfactual contrast between value corruption and structural perturbation indicates the signal tracks abstract topology rather than identifier values, and replicates under an independent, non-substring oracle. The non-positional component replicates on three further interactive multi-hop benchmarks and attenuates as call order alone becomes a sufficient proxy for dependency, vanishing in single-shot planning. Per-layer activation patching shifts the probe at a later, non-patched boundary, evidence that the representation propagates rather than passively reads out, though the realised tool call does not move. To our knowledge this is the first structural probe of an LLM agent's runtime tool-call dependency graph. Our claims concern representation, not behavioural control, and span two model families and one primary domain.
arXiv:2604.20022v3 Announce Type: replace
Abstract: Large language models (LLMs) are increasingly used for conversational clinical decision support, yet they conflate next token prediction with probabilistic decision making. We argue that this conflation reflects an architectural limitation: such systems lack explicit posterior tracking, controllable abstention thresholds, and auditable reasoning chains. We introduce MoBayes, a Modular Bayesian dialogue framework that separates reasoning from language. The LLM acts only as a language interface, parsing patient conversation into structured observations, while a Bayesian module performs probabilistic inference over these observations to update posteriors, select follow-up questions via expected-information-gain and determine when to stop or defer through calibrated decision thresholds. This design enables explicit posterior tracking, controllable selective decision-making, and replaceable population-specific statistical backends without retraining the language model. Across empirical and LLM-generated knowledge bases, MoBayes outperforms standalone frontier LLM doctors, including matched model-family comparisons where inexpensive sensor models paired with MoBayes exceed larger autonomous models at lower cost. The advantage persists under adversarial patient communication styles and across varying diagnostic scenarios. These results suggest that reliable conversational clinical decision support systems should separate probabilistic reasoning from language generation rather than scaling model size alone. Code is available at https://anonymous.4open.science/r/MoBayes/
arXiv:2605.25045v1 Announce Type: new
Abstract: Time series research is moving beyond fixed forecasting benchmarks toward realistic tasks that combine prediction, contextual reasoning, tool use, and structured decision support. Most benchmarks are built around clean data and short evaluation loops; agents alone may miss temporal constraints, evidence checks, or review before finalizing outputs. We first formalize next-generation time series tasks as three-component tuples consisting of a task file, a workspace, and a validation interface. We then present AION, a time series harness built from six component groups: agents, skills, rules, memory, evaluation, and protocols. In this harness, we use three design principles: temporal grounding, temporal knowledge-grounded reasoning, and reliability mechanisms such as post-experiment analysis and layered review. A Kaggle Store Sales case study shows that the harness produces more detailed process traces, more artifacts, and more review steps than the same base agent operating in OpenCode direct build mode. Taken together, these results argue for a paradigm shift from fixed tasks to realistic ones under real-world constraints.
arXiv:2605.25939v1 Announce Type: new
Abstract: We here study whether training biases can make hidden neurons specialize in minimal one-hidden-layer MLPs, and whether such specialization improves prototype-based reconstruction of the training dataset from the learned weights. We consider Gaussianactivation MLPs of width equal to dataset size and compare three structural losses that respectively encourage coverage of the training samples, separation between neuron-induced prototypes, and low overlap of hidden responses, against the standard fitting baseline. Experiments on uniformly sampled one-dimensional datasets show a stable pattern from N = 3 to N = 100 across 480 controlled runs. Coverage regularization gives the lowest mean reconstruction error at every tested size and raises the prototype-usage specialization ratio relative to the standard baseline, while separation has mixed effects and overlap penalties are systematically harmful. We show that the harm is not an optimization failure: overlap-active approaches fit the data as well as overlap-free ones but route the optimizer to a degenerate equilibrium in which prototype centers are pushed outside the convex hull of the training inputs. Coverage cannot reward this expulsion and acts as an attractor: separation admits it only at large temperature and overlap admits it at the nominal hyperparameter choice. A direct {\tau}-sweep on the separation-only mask and a prototype-position visualization at N = 100 confirm the mechanism. The findings yield a simple design principle for prototype-recoverability-aware training: every repulsive structural loss must be compensated by a compatible attractor, or it will collapse the latent geometry it was meant to refine.
arXiv:2601.03327v3 Announce Type: replace
Abstract: Wildfires are highly imbalanced natural hazards in both space and severity, making the prediction of extreme events particularly challenging. In this work, we introduce the first ordinal classification framework for forecasting wildfire severity levels directly aligned with operational decision-making in France. Our study investigates the influence of loss-function design on the ability of neural models to predict rare yet critical high-severity fire occurrences. We compare standard cross-entropy with several ordinal-aware objectives, including the proposed probabilistic TDeGPD loss derived from a truncated discrete exponentiated Generalized Pareto Distribution. Through extensive benchmarking over multiple architectures and real operational data, we show that ordinal supervision substantially improves model performance over conventional approaches. In particular, the Weighted Kappa Loss (WKLoss) achieves the best overall results, with more than +0.1 IoU (Intersection Over Union) gain on the most extreme severity classes while maintaining competitive calibration quality. However, performance remains limited for the rarest events due to their extremely low representation in the dataset. These findings highlight the importance of integrating both severity ordering, data imbalance considerations, and seasonality risk into wildfire forecasting systems. Future work will focus on incorporating seasonal dynamics and uncertainty information into training to further improve the reliability of extreme-event prediction.
arXiv:2605.25931v1 Announce Type: new
Abstract: We systematically investigate all 25 public ARC-AGI-3 games and find that every one is reachable through non-intelligent strategies: 10 in a single blind step, 5 after one probing action, 1 via repeated ACTION1 presses, 1 via diverse exploration, and 8 via single repeated actions with sufficient budget (50-200 steps). A library-level null-coordinate vulnerability additionally bypasses 18 games in 1 step. This benchmark critique implies the public evaluation set cannot discriminate intelligent exploration from trivial heuristics - the private 55-game evaluation is the only genuine intelligence test. Against this backdrop, we present AERA (Adaptive Epistemic Reasoning Agent), a three-phase (EXPLORE / VERIFY / PLAN) agent achieving RHAE=0.2116 (4/25 solved) on these 25 games with Qwen2.5-0.5B, while random and no-explore baselines score 0.0000. We formalise AERA through a Speed--Depth trade-off framework: under a convexity assumption (proved for a class of environments in the Appendix), RHAE's quadratic form emerges as a second-order penalty for deviating from the Pareto frontier between action efficiency and information gain. Contributions: (i) a benchmark validity analysis showing that current interactive reasoning benchmarks fail to measure the exploration they claim to require, and (ii) the EXPLORE-before-PLAN framework and model-capability x exploration interaction. The linked code track entry achieves RHAE=0.30 on the full 55-game private evaluation. Code: CC0.
arXiv:2605.25920v1 Announce Type: new
Abstract: While large language models (LLMs) augmented with agentic search capabilities show promise for legal reasoning, they overlook a fundamental constraint that applicable law must match the temporal context of each case, as retroactive application of statutes violates core legal principles and leads to erroneous conclusions. Our observations reveal that current legal LLMs suffer from temporal bias anchored to their training cutoff, while search agents rarely incorporate temporal constraints into queries, and that web search alone cannot provide the precise statute and precedent citations that legal reasoning demands. To address these challenges, we propose LegalSearch-R1, an end-to-end reinforcement learning framework that pairs local statute RAG for precise article matching with online web search for broader legal knowledge, trained on temporally-indexed data spanning multiple amendment periods to enforce temporal consistency. Extensive experiments on our benchmark covering 13 legal tasks demonstrate that our 7B-parameter agent outperforms state-of-the-art deep research frameworks and specialized legal LLMs by 12.9% to 29.8%, surpasses baselines by 57.7% to 80.3% on temporal consistency, and exhibits robust out-of-domain generalization. The code and data are available at https://github.com/AlexFanw/LegalSearch-R1.
arXiv:2605.24829v1 Announce Type: new
Abstract: Full-potential electronic structure calculations for periodic systems retain the Coulomb singularity at the nuclei, which induces cusp behavior of the orbitals near the nuclei while leaving the interstitial region smooth. This multiscale regularity motivates a hybrid discretization framework that combines localized real-space discretizations in atomic patches with plane waves in the interstitial region. In this work, we employ tensor-product B-splines in the atomic patches and couple the two approximation spaces through a symmetric interior penalty discontinuous Galerkin formulation, yielding the isogeometric-plane wave (IGA-PW) method. To efficiently evaluate the restricted plane wave integrals, we develop a combined fast Fourier transform and Chebyshev correction strategy. We also construct a trace-block DG preconditioner to alleviate the conditioning deterioration caused by the SIPG penalty terms and accelerate the iterative eigensolver. In addition, we establish {\it a priori} error estimates for the corresponding linear eigenvalue problem, showing algebraic convergence near the nuclei and superalgebraic convergence in the interstitial region. Numerical experiments demonstrate the accuracy and efficiency of the proposed method.
arXiv:2605.25909v1 Announce Type: new
Abstract: Reconstructing and predicting dynamic 3D scenes from multi-view videos is a foundational task for robotics, AR/VR, and digital twins. Recent physics-informed Gaussian Splatting methods achieve impressive future frame extrapolation but lack semantic awareness and suffer from large computational overhead. We introduce $\textbf{R5DGS}$, a framework that augments a physics-driven 4D Gaussian representation with compact Identity Encoding vectors, enabling precise Gaussian-to-object association. By constructing an offline CLIP-based object lookup table, we support open-vocabulary text prompting to retrieve and render object-specific Gaussians across arbitrary timestamps and viewpoints. Furthermore, we propose a rigid-body inference constraint that predicts and integrates physical dynamics exclusively for object centroids, propagating motion to associated Gaussians via relative transformations. This optimization yields a 11 FPS speedup during extrapolation without compromising trajectories plausibility.
arXiv:2605.25313v1 Announce Type: new
Abstract: World models for partially observed environments must imagine multiple compatible hidden futures and steer between them under counterfactual actions. Joint Embedding Predictive Architectures (JEPAs) do this in latent space, but a vector-valued latent has no internal structure for carrying the belief over hidden continuations through blind rollout. We introduce the Unitary World Model JEPA (UWM-JEPA), a JEPA world model with a density-matrix latent on a joint system-environment space and a learned unitary predictor. The construction preserves the joint-state spectrum exactly during rollout, so the predictor itself cannot dissipate the represented uncertainty.
On a hidden-velocity indicator task requiring five-step forward simulation under a given action sequence with the target observation masked, UWM-JEPA reaches 0.77 accuracy and degrades monotonically as actions are perturbed; a parameter-matched LSTM-JEPA trained under the same counterfactual-target objective and action head collapses to majority-class accuracy (0.53) under every action condition. Under blind rollout, UWM-JEPA loses fewer than ten points of probe R^2 at short horizons while vector-latent baselines lose forty-one and sixty-eight; both nevertheless tie on a held-out context probe, locating the separation in the predictor rather than the encoder. Action sensitivity itself requires training against counterfactual rather than teacher-forced targets, a finding that applies beyond the unitary parameterisation. For JEPA world models to imagine under partial observability, latent geometry and predictor dynamics matter, not frozen context-encoding capacity alone.
arXiv:2602.22631v2 Announce Type: replace
Abstract: Neural networks are increasingly deployed in scientific, safety critical, and mission critical pipelines, yet verification and analysis are often performed outside the programming environment that defines and runs the model. This creates a semantic gap between the executed network and the analyzed artifact: guarantees can depend on implicit conventions about operator semantics, tensor layouts, preprocessing, floating-point behavior, graph transformations, accelerated kernels, and external certificates. We present TorchLean, a unified framework for formalizing, executing, and verifying neural networks in Lean 4. TorchLean treats learned models as executable programs and mathematical objects with a shared semantics for computation, verification, and theorem proving. The framework provides a PyTorch style API for typed tensors, layers, objectives, optimizers, automatic differentiation, and graph programs, with eager and compiled execution paths that lower to a common computation-graph representation. TorchLean supports exact and finite-precision tensor semantics, verified reverse-mode differentiation, interval and affine bound propagation, CROWN/LiRPA style certificate checking, import/export workflows, and CUDA-backed execution through explicit FFI boundaries. It also includes semantic layers for attention and FlashAttention, state-space sequence models, diffusion and sampling processes, probability kernels, reinforcement-learning objectives and Markov decision processes, and self-supervised objectives such as masked autoencoding, JEPA-style predictive views, and variance/correlation-based anti-collapse losses. Together, these components provide a semantic foundation for verified machine learning, where executable neural network artifacts, verification procedures, runtime boundaries, and mathematical claims can be stated and related inside one theorem-proving environment.
arXiv:2605.25895v1 Announce Type: new
Abstract: Excess molar enthalpies ($H_{\text{m}}^{\text{E}}$) for iodobenzene, or 1-iodonaphthalene + heptane, + decane, + dodecane, or + tetradecane mixtures at 298.15 K and 93 kPa have been measured using a Tian-Calvet micro-calorimeter. The values of $H_{\text{m}}^{\text{E}}$ are positive and indicate that interactions between like molecules are prevalent. In contrast, our previous results on excess molar volumes ($V_{\text{m}}^{\text{E}}$) are negative for the systems C$_6$H$_5$I + heptane, or 1-iodonaphthalene + n-alkane, which reveal the existence of large structural effects in such solutions. This set of measurements has been used to determine isochoric excess molar internal energies ($U_{V\text{m}}^{\text{E}}$). In the range of n-alkanes considered (n is the number of C atoms of the alkane), values of $U_{V\text{m}}^{\text{E}}$ at equimolar composition decrease from n = 7 to n = 10 and then slightly increase for systems with C$_6$H$_5$I, while decrease slowly for mixtures with 1-iodonaphthalene. These trends fit well with the patterns observed for other alkane mixtures containing cyclic molecules. Dispersive interactions are dominant and those between aromatic molecules with a given halogen atom become stronger when the size of this atom increases due to the corresponding increase of molecular polarizability. The mixtures were studied using the DISQUAC and Flory models. The latter was also applied to n-alkane solutions with C$_6$H$_5$F, or 1-methylnaphthalene. Both theories describe accurately the $H_{\text{m}}^{\text{E}}$ data. In terms of the Flory model, this means that the random mixing hypothesis is largely achieved. On the other hand, the theory overestimates the interactional contribution to $V_{\text{m}}^{\text{E}}$, particularly for systems with C$_6$H$_5$X (X = F, I).
arXiv:2605.14334v2 Announce Type: replace
Abstract: We study the problem of secure joint source-channel coding for multimodal semantic sources transmitted over noisy wiretap channels. The source model consists of $m$ modalities (e.g., image, audio, and sensor data), all represented as random variables. The encoder observes independent and identically distributed samples of an arbitrary non-empty subset of modalities. The samples are encoded and transmitted over a discrete memoryless wiretap channel. The legitimate receiver reconstructs all modalities. We extend the rate-distortion-perception problem formulation to multimodal sources. We establish converse and achievability bounds on the fundamental limits of transmission rate, fidelity, and secrecy, under per-modality distortion and perception constraints, and per-subset equivocation constraints. We show that the fundamental limit for secrecy consists of three operationally distinct components: the level of compression, the secret key rate, and the statistics of the wiretap channel.
arXiv:2510.00479v3 Announce Type: replace
Abstract: We numerically investigate the feasibility and limits of jointly estimating flow fields and unknown particle properties (e.g., position, size, and density) from Lagrangian particle tracking (LPT) data. LPT offers time-resolved, volumetric measurements of particle trajectories, which are markers of the carrier fluid motion. However, experimental tracks are spatially sparse and potentially noisy, and the problem of reconstructing flow fields may be further complicated by inertial particle transport, such that particle slip velocities must be determined to access the velocity field of the carrier fluid. To address this problem, we develop a data assimilation framework that couples an Eulerian representation of the flow with Lagrangian particle models, enabling the simultaneous inference of carrier fields and particle properties under the governing equations of disperse multiphase flow. We show that flow fields and particle properties can be jointly estimated in three representative regimes: (1) In a turbulent boundary layer with noisy tracer tracks (St to 0), flow fields and true particle positions are jointly estimated, which amounts to a physics-informed particle tracking problem; (2) in homogeneous isotropic turbulence seeded with inertial particles (St ~ 1-5), we demonstrate simultaneous recovery of flow states and particle diameters, showing the feasibility of implicit particle characterization; and (3) in a compressible, shock-dominated flow, we report the first joint reconstructions of velocity, pressure, density, and inertial particle properties (diameter and density), highlighting both the potential and certain limits of joint estimation in supersonic regimes. A systematic sensitivity study reveals how the seeding density, noise level, and Stokes number govern reconstruction accuracy for our method.
arXiv:2605.25126v1 Announce Type: cross
Abstract: For a full-rank integral lattice $\mathcal{L}\subset\mathbb{R}^n$, Regev and Stephens-Davidowitz proved that \[N_{=k}(\mathcal{L}):=|\{y\in\mathcal{L}:\lVert y\rVert^2=k\}|\le 2\binom{n+2k-2}{2k-1}.\] We classify the equality cases. For $n\ge2$, equality holds if and only if either $k=1$ and $\mathcal{L}\cong\mathbb{Z}^n$, or $n=8$, $k=2$, and $\mathcal{L}\cong E_8$. For $n=1$, equality holds exactly when $\mathcal{L}$ represents $k$.
The proof shows that equality is rigid. Saturation of the shell bound forces the normalized norm-$k$ shell to be an antipodal tight spherical $(4k-1)$-design. The associated Delsarte--Goethals--Seidel annihilator polynomial gives an arithmetic root condition, which isolates $E_8$ at $k=2$, rules out $k=3$, and combines with the Bannai--Damerell/Bannai theorem and an elementary circle argument to exclude all remaining cases in dimension at least $2$.
arXiv:2605.25679v1 Announce Type: new
Abstract: Accurate prediction of three-dimensional (3D) wind fields over complex mountainous terrain is essential for renewable energy deployment and regional weather modeling. Traditional computational fluid dynamics (CFD) simulations face two fundamental bottlenecks: expert-intensive mesh generation around irregular topography, and iterative solvers that require hours to days even on high-performance clusters. Recent neural operator approaches accelerate inference, but typically fail to resolve the sharp, localized velocity gradients induced by complex terrain features. Here, we present a transformer-based dual-attention neural-operator framework for 3D wind field prediction over complex mountainous terrain, and validate its effectiveness through two instantiations on representative point-based (mesh-free) and graph-based neural-operator architectures, namely Patch-solver and Patch-GTO. Trained on a large CFD-generated dataset spanning diverse terrain geometries and inflow conditions, the framework enables rapid prediction of steady-state wind field while maintaining competitive accuracy. It also demonstrates robust zero-shot transfer to real-world mountainous sites across several diverse locations, outperforming existing neural operator baselines by 10% in relative error. We further verify that incorporating sparse observational data (1% spatial coverage) reduces prediction error by 16.89% relative to the corresponding model without sparse data input and by 32.75% relative to advanced neural operator baselines on unseen terrains. This framework establishes a generalizable computational paradigm across domains, promising to be a real-time tool for wind resource assessment over complex mountainous terrain and related atmosphere-surface interaction studies.
arXiv:2605.25881v1 Announce Type: new
Abstract: Dielectric metasurfaces offer a path to high-efficiency optical components at the sub-wavelength scale. In this work, we utilize the wavelength-dependent birefringence of TiO2 nanopillars to create transmissive color filters with switchable spectral responses. We demonstrate that the same physical nanostructure array can function as either an additive (band-pass) or a subtractive (band-stop) filter solely by rotating an external polarizer. The metasurface was fabricated via a modified damascene process, enabling high-aspect-ratio structures with spatial resolutions down to 250 nm. The filters achieve transmission efficiencies of 70% tailorable across the entire visible spectrum, avoiding the ohmic losses typical of plasmonic alternatives. This approach provides a tunable, high-efficiency platform offering a versatile and compact solution for next-generation high-resolution imaging systems.
arXiv:2603.16100v2 Announce Type: replace
Abstract: Recent research suggested that the embeddings produced by CLIP-like contrastive language-image training are suboptimal for image-only tasks. The main theory is that the inter-modal (language-image) alignment loss ignores intra-modal (image-image) alignment, leading to poorly calibrated distances between images. In this study, we question this intra-modal misalignment hypothesis. We reexamine its foundational theoretical argument, the indicators used to support it, and the performance metrics affected. For the theoretical argument, we demonstrate that there are no such supposed degrees of freedom for image embedding distances. For the empirical measures, our findings reveal they yield similar results for language-image trained models (CLIP, SigLIP) and image-image trained models (DINO, SigLIP2). This indicates the observed phenomena do not stem from a misalignment specific to the former. Experiments on the commonly studied intra-modal tasks retrieval and few-shot classification confirm that addressing task ambiguity, not supposed misalignment, is key for best results.