Systems and Control
Characterization and Monitoring of Nonlinear Dynamics and Chaos in Complex Physiological Systems
Hui Yang, Yun Chen and Fabio Leonelli
Nonlinear dynamics arise whenever multifarious entities of a system cooperate, compete, or interfere. For example, cardiovascular system involves a great level of complexity. Multi-lead ECG signals are generated through orchestrated depolarization and repolarization of
cells and manifest significant nonlinear dynamics. Nonlinear dynamical systems defy understanding based on the traditional reductionist's approach, in which one attempts to understand a system's behavior by combining all constituent parts that have been analyzed separately. In order to cope with system complexity, modern healthcare systems are investing in advanced physiological sensing and patient monitoring, thereby giving rise to big data. Realizing the full potential of big data for healthcare intelligence requires fundamentally new methodologies to harness and exploit complexity. However, available nonlinear dynamics techniques are either not concerned with healthcare analytical objectives or fail to effectively analyze big data to extract useful information for improving healthcare services. There is an urgent need to develop analytical methodologies that fully exploit the underlying nonlinear dynamics in physiological systems for advancing healthcare services with exceptional features such as personalization, responsiveness, and superior quality. This chapter presents some theoretical developments and tools to advance the applications of nonlinear dynamics principles in health care. Specifically, we focus on sensor-based characterization and modeling of nonlinear dynamics (i.e., multifractal analysis and multiscale recurrence quantification). Then, current developments and applications of these methodologies are examined for characterizing and exploiting heart rate variability and space-time ECG signals.
Read the full abstract ▾
Computer Vision
REALIS: A Curated Dataset for Studying the Challenges of AI Image Detection
Aleksandr Gushchin, Khaled Abud, Georgii Bychkov et al. (+5 more)
AI-generated image detectors are often evaluated on
benchmarks where real and synthetic images differ in content, quality, or generation artifacts, allowing models to rely on
dataset-specific cues and fail on unfamiliar generators or processed images. Existing
datasets provide limited support for evaluating these challenges jointly across diverse visual content. We introduce REALIS, a
dataset of 1.43 million real and synthetic images generated by 42 modern text-to-image models, including the latest proprietary systems such as Nano Banana 2. REALIS combines
prompts derived from real images, quality filtering, and stratified sampling to reduce class-specific shortcuts while preserving content diversity. We further introduce REALIS-Expert, a stress-test subset for high-quality synthetic images, where real and generated samples are selected with closely matched semantic and visual characteristics. We also propose a robustness protocol covering 35 transformations at five severity levels to analyze detector behavior under image processing. Based on REALIS, our
benchmark evaluates pretrained detectors, fine-tuned models, and zero-shot
vision-
language models under generator and post-processing shifts. On the hardest processed split, the best pretrained conventional detector achieves 0.550 ROC-AUC, compared with 0.752 for the best REALIS-trained detector. REALIS provides a unified framework for measuring and improving the reliability of AI-image detectors under conditions that better reflect real-world use.
Read the full abstract ▾
All authors Aleksandr Gushchin, Khaled Abud, Georgii Bychkov, Ekaterina Shumitskaya, Artem Filippov, Sergey Lavrushkin, Dmitriy S. Vatolin, Anastasia Antsiferova
Algebraic Topology
A data structure for quotient flag complexes
Konstantin Sorokin, Aleksandr Levin, Maxim Beketov et al. (+1 more)
Vietoris-Rips filtrations, which are standard in topological data analysis, are flag complexes, and a simplex tree stores these without any attaching data. In this paper we ask what survives of this economy when a flag complex $K$ is divided by a subcomplex $A$, each connected component of $A$ being crushed to a point. Such a quotient is a CW complex whose
cells are the simplices of $K\setminus A$, but their attaching maps are no longer implicit. We show that for flag $K$ the face order of the quotient is strictly graded exactly when $A$ is flag, and that the surviving labelled
cells are determined by those of dimension at most 3. For $m$-flag pairs the threshold is $2m+1$, and it drops to $m+2$ when $K$ is flag. The prescribed
cells form a regular CW decomposition only when $A$ is full in $K$. These results justify the QF-tree: a
cell table that stores, for each surviving simplex, its ordered list of $d+1$ facets with collapsed facets flagged, indexed by a trie of quotient-vertex words. For bounded dimension its size is linear in the number of surviving simplices plus the retained provenance, and we derive and verify a simple formula for the collapsed fraction above which it is smaller than the homotopy-equivalent cone model. Because a collapse changes the attaching data only on the closed
star of $A$, the QF-tree can also be applied locally inside a simplex tree. For a ball-shaped $A$ in the sampled Vietoris–Rips regime the closed
star is a thin shell, and the median compact budget is below the cone model at every sampled radius. An accompanying library, modelled on Gudhi, implements the QF-tree, its local variant, an editable layer with local quotient updates, gluing, disc attachment, induced maps, cup products, fundamental-group presentations and zigzag persistence, and provided experiments separate the cost of maintaining a quotient from the cost of the
algebra computed on it.
Read the full abstract ▾
All authors Konstantin Sorokin, Aleksandr Levin, Maxim Beketov, Anton Ayzenberg
Social and Information Networks
Plan-to-Synthesis: Cross-City Human Mobility Generation via Semantic Latent Flow Matching
Zhoufu Wang, Baoshen Guo, Zhiqing Hong et al. (+6 more)
Human mobility generation aims to synthesize realistic point-of-interest (POI) visitation trajectories and has become an important tool for travel behavior modeling, transportation management, and urban planning. Existing
diffusion-based methods achieve high fidelity but require per-city generation, given the inherent heterogeneity of geospatial locations and POI categories, while large
language model-based methods generalize across cities but remain too costly at scale, especially for long-horizon trajectory generation. To address this, we propose SeMoFlow, a Semantic human Mobility generation framework based on latent Flow matching. We first encode heterogeneous POIs from different cities into a shared cross-city representation space via hierarchical Semantic IDs, where shared prefixes capture transferable semantics, and successive codes progressively refine the representation toward individual POIs. Building on the semantic IDs, SeMoFlow follows a plan-to-synthesis hierarchical generation paradigm, in which an autoregressive planner generates coarse-grained semantic and recurrence patterns, and a flow matching realizer synthesizes fine-grained suffix latents. The generated latents are subsequently decoded and grounded to concrete POIs. Extensive experiments on large-scale multi-city
datasets show that SeMoFlow achieves higher trajectory fidelity than existing baselines, preserves city-specific mobility motifs, and supports both joint multi-city generation and effective cross-city transfer. The code is available at \href{https://github.com/urban-mobility-generation/SeMoFlow}{https://github.com/urban-mobility-generation/SeMoFlow}.
Read the full abstract ▾
All authors Zhoufu Wang, Baoshen Guo, Zhiqing Hong, Junyi Li, Kailai Sun, Heye Huang, Alok Prakash, Shenhao Wang, Jinhua Zhao
Artificial Intelligence
SkillVine: Agent Skill Evolution via Branching Exploration
Kaiwei Liu, Jiqian Dong, Liran Dong et al. (+7 more)
Agent skills encapsulate reusable procedural knowledge that enables LLM
agents to perform tasks, and they can be improved automatically using trajectories from interactions with the environment. This is the classic problem of skill
evolution. Existing approaches predominately follow a linear
evolution paradigm, in which updates are sequentially applied to the latest skill-library version. As a result, they inevitably fall into local optima, leaving many promising
evolution paths unexplored. We propose SkillVine, an automatic skill-
evolution framework that formulates skill
evolution as a graph search problem and employs a branching exploration strategy. Equipped with a trunk-branch collaborative searching mechanism, an intelligent parent-node selector, and an adaptive-granularity update rule, SkillVine achieves a balance between exploration and exploitation. We evaluate SkillVine on 5
benchmarks with two LLMs. Results show that SkillVine discovers better skill-library versions along branches than along the linear trunk and achieves the best test performance in nine of ten
benchmark-model combinations.
Read the full abstract ▾
All authors Kaiwei Liu, Jiqian Dong, Liran Dong, Shuai Mao, Mingming Zhao, Bufang Yang, Jie Chuai, Zhitang Chen, Guoliang Xing, Zhenyu Yan
Machine Learning (Statistics)
Domain Adaptation with Target Information via Doubly-Anchored Distributionally Robust Optimization
David Kepplinger and Anand N. Vidyashankar
Domain Adaptation (DA) often lacks worst-case guarantees, while Distributionally Robust Optimization (DRO) based only on source data centers its ambiguity set at the source law and ignores available target structure. To bridge this gap, we introduce a doubly-anchored DRO framework whose ambiguity set is the intersection of $\phi$-divergence balls centered at the source law and a source-completed target reference law, the latter pairing the target
covariate law with the source conditional law. We derive dual-induced
adversarial bridge geometries for symmetric and asymmetric divergence pairings, notably introducing a Kullback–Leibler/squared-Hellinger (KL/HD) bridge. This asymmetric formulation yields a Lambert-$W$ geometry in which source-side exponential risk tilting and target-side Hellinger stabilization enter through distinct terms, attenuating, but not bounding, the effect of large likelihood ratios. Furthermore, without imposing
covariate shift, we establish finite-sample generalization bounds for the minimizer of a structural, loss-agnostic density-bridge risk under bounded-overlap conditions; these bounds do not apply directly to the loss-aware DRO min–max
estimator. We translate our framework into a bridge-weighted Nadaraya–Watson
estimator, proving uniform consistency for the source regression function and pointwise asymptotic normality, with target recovery when the source and target regression functions coincide, as under
covariate shift. Finally, an empirical evaluation on a domain-shifted Fashion-MNIST
dataset illustrates the finite-sample stability of the asymmetric KL/HD bridge under severe synthetic target-
covariate corruption.
Read the full abstract ▾
Optimization and Control
Revisiting AdaGrad in Stochastic Convex Optimization: Last Iterates, High Probability, and Lower Bounds
Weiming Ou and Xiao Wang
AdaGrad and AdaGrad-Norm are widely used adaptive methods, but their precise behavior in
stochastic convex optimization remains less understood. We first show that AdaGrad-Norm and AdaGrad do not admit any universal last-iterate rate, even under sub-Gaussian noise and bounded iterates. We then prove that bounded variance alone is too weak: even with bounded iterates, it cannot yield high-probability average-iterate rates, and without bounded iterates it may not even guarantee convergence in expectation. Besides, we construct tight $\Omega(\log T/\sqrt T)$ lower bounds for both AdaGrad-Norm and AdaGrad under sub-Gaussian noise, showing that $\log T$ in existing average-iterate upper bounds is unavoidable. Finally, we show that this $\log T$ loss disappears once bounded-iterate condition is imposed: under a general ABC condition, both methods achieve rates ${O}(1/\sqrt{T})$.
Read the full abstract ▾
Data Structures and Algorithms
Single-Exponential Algorithms for Directed Feedback Vertex Set on Planar Digraphs
Daniel Lokshtanov, Saket Saurabh and Jie Xue
We consider Directed Feedback Vertex Set on planar digraphs, parameterized by the solution size $k$. We give a randomized
algorithm with one-sided error running in time $(2+\sqrt5)^k n^{O(1)}= 4.24^k n^{O(1)}$, and a deterministic
algorithm running in time $8.04^k n^{O(1)}$. Both
algorithms use
polynomial space. To the best of our knowledge, these are the first single-exponential fixed-parameter
algorithms for Directed Feedback Vertex Set on planar digraphs. This contrasts with general digraphs, where the best known
algorithms run in time $2^{O(k\log k)}(n+m)$, and whether a $2^{o(k\log k)}n^{O(1)}$-time
algorithm exists remains a major open problem. Our main tool is an exact Euler-type counting identity for plane digraphs. It shows that every small solution must carry a large share of the vertices whose in- and out-arcs alternate in the
embedding, while solutions avoiding such vertices can be computed by reducing to Directed Feedback Arc Set, which is known to be solvable in
polynomial time on planar digraphs via the Lucchesi-Younger
theorem.
Read the full abstract ▾
Machine Learning
Distributionally Robust Average-Reward Reinforcement Learning: Finite-Sample Guarantees under Weak Communication
Chenyu Lu, Zijun Chen and Nian Si
We study distributionally robust reinforcement learning (DR-RL) in the average-reward setting under weak communication. Our main result provides finite-sample guarantees for estimating the robust optimal average reward and learning a near-optimal policy, covering both SA-rectangular and S-rectangular structures with divergence-based and distance-based uncertainty sets. Specifically, for Kullback–Leibler and $f_k$-divergence balls, we establish explicit radius conditions under which the robust average-reward Bellman equation admits a constant-gain solution, while for total variation and Wasserstein balls, any positive radius suffices without requiring the nominal MDP to be weakly communicating. Our
algorithm is prior-knowledge-free and achieves sample complexities of $\widetilde O(|\mathcal{S}||\mathcal{A}|p_{\wedge}^{-1}\operatorname{Span}^{2}(u_{\delta}^{\ast})\epsilon^{-2})$ for estimating the robust optimal average reward and $\widetilde O(|\mathcal{S}||\mathcal{A}|p_{\wedge}^{-2}\operatorname{Span}^{2}(u_{\delta}^{\ast})\epsilon^{-2})$ for learning an $\epsilon$-optimal policy. Here, $p_{\wedge}$ is the smallest positive nominal transition probability and $u_{\delta}^{\ast}$ is a robust optimal bias function. We further provide an almost-tight explicit upper bound on $\operatorname{Span}(u_{\delta}^{\ast})$. Finally, we validate the predicted $n^{-1/2}$ convergence rate through numerical experiments.
Read the full abstract ▾
High Energy Astrophysical Phenomena
Pulse profiles of tall accretion columns in X-ray pulsars
Bagration Megrelishvili and Pavel Abolmasov
Accreting X-ray
pulsars (XRPs) are strongly magnetized
neutron stars (NSs) distinguished by a periodic variability pattern related to the rotation of their magnetospheres (pulsations). Many of these objects show complex asymmetric pulse profiles that are difficult to explain in the framework of existing models. We calculate the pulsations of X-ray radiation from a magnetized NS in the regime of supercritical
accretion, when the radiation shock wave is high above the NS surface, and the observed radiation comes from two optically thick
accretion columns located downstream of the shocks. We show that the
accretion column beam pattern calculated from the first principles has a potential of reproducing the variety of XRP pulse shapes. The surfaces of the columns are assumed to radiate locally as blackbodies with the temperature distribution following the solution of Basko and Sunyaev (1976). The adopted geometry of the field is an inclined dipole. It is taken into account that the two
accretion flows above the shocks attenuate and scatter the radiation of the
accretion columns. Using the ray-tracing technique, we obtain a vast variety of pulse shapes controlled by the mass
accretion rate,
magnetic angle, the direction towards the observer, and the parameters that determine the particular set of
magnetic field lines being loaded with accreting matter. We propose a universal dimensionless parameter describing the asymmetry of a pulse quantitatively. The model naturally produces asymmetric pulse profiles observed in many XRPs. Its applicability, however, is limited to the brightest objects like pulsating ultraluminous Xray sources and strong
flares of
Galactic XRPs.
Read the full abstract ▾
Econometrics
Minimax Choice of Projection Geometry under Linear Inequality Constraints
Joachim Freyberger and Julius Kappenberg
Economic theory frequently implies linear inequality restrictions on parameters or functions of interest. A common way to impose such restrictions is to project an unrestricted
estimator onto the feasible set. Projection
estimators arise naturally from constrained least squares, instrumental variables, generalized method of moments, maximum likelihood, and related extremum procedures. When the sampling covariance, loss function, and projection criterion induce different geometries, the choice of projection geometry can substantially affect risk. We study this choice in a fixed-dimensional local Gaussian experiment under quadratic loss. At exact-boundary configurations where only one maintained inequality binds, inverse-covariance projection is pointwise optimal. When at most two inequalities are locally relevant, it weakly improves on the unrestricted
estimator throughout the corresponding local experiment and is minimax over exact-boundary configurations. For an arbitrary number of inequalities, we provide a sufficient condition for boundary minimaxity, but show by counterexample that inverse-covariance projection need not be boundary minimax once three inequalities can bind. Motivated by these results, we propose selecting the projection geometry to minimize worst-case boundary risk subject to a local no-harm condition relative to unrestricted estimation. We develop a feasible implementation and study its finite-sample performance in simulations and an application to gasoline demand.
Read the full abstract ▾
Earth and Planetary Astrophysics
Pre-accretional irradiation of comet 67P/Churyumov-Gerasimenko constituents
Inès Sansberro, Nicolas Fray, Hervé Cottin et al. (+11 more)
Comets are thought to have largely preserved the
primordial material from which they formed. Therefore, cometary studies provide important clues about the processes involved in the formation of the
Solar System. During the last weeks of ESA's Rosetta mission, the COSIMA (COmetary Secondary Ion Mass Analyser) instrument performed a series of measurements, which involved long-term sputtering of
dust particles captured from comet 67P/Churyumov-Gerasimenko. These post-sputtering analyses revealed the subsurface composition of 8 particles, which we compare here with surfaces composition measured on 59 un-sputtered particle fragments for 9 elements. We found that the analysed surfaces are strongly depleted in metals, such as Mg and Fe, compared to
solar abundances, while the subsurface material is consistent with
solar values. Because the
dust captured by COSIMA fragmented upon impact with the collection targets, this metal depletion cannot reside exclusively on the outer surface of the bulk particles. Instead, these depletion layers must coat the surface of smaller, unfragmented mineral subunits. Such depletion layers, which are often called 'rims', are common at the surface of small regolith
dust particles found on airless bodies in our
solar system and are attributed to space weathering. We suggest that these mineral cometary subunits experienced pre-accretional
solar wind irradiation in the inner regions of the
solar nebula during the early stages of
Solar System formation. The small irradiated
dust constituents must then have been transported to the outer regions of the
solar nebula, where they mixed with ices to form the
nucleus of the comet.
Read the full abstract ▾
All authors Inès Sansberro, Nicolas Fray, Hervé Cottin, Cécile Engrand, Donia Baklouti, Christelle Briois, Anaïs Bardyn, Henning Fischer, Klaus Hornung, John Paquette, Jouni Ryno, Olivier J. Stenzel, Laurent Thirkell, Martin Hilchenbach
Category Theory
Relative dualizability and the cobordism hypothesis for defects
William Stewart
This paper compares two candidate notions of morphism in the setting of fully local topological field theory. A first description can be formulated in terms of defects of codimension $k$ separating a pair of defects of codimension $(k-1)$. Lurie's cobordism hypothesis with singularities identifies such defects with suitably dualizable $k$-morphisms in the target category. A second description can be formulated in terms of symmetric monoidal (op)lax natural transformations between the truncations of the source and target theories. We show that these two descriptions agree. The comparison rests on a criterion for when a $k$-morphism determines a codimension-$k$ defect, expressed as a one-sided iterated adjunctibility condition. Assuming the ordinary cobordism hypothesis, this comparison provides a reformulation of the cobordism hypothesis for defects in terms of systems of oplax natural transformations. We also apply the criterion to the higher Morita category, where it yields a reducibility
theorem for modules over $E_n$-
algebras: a module over an $(n+1)$-dualizable $E_n$-
algebra $A$ is $n$-dualizable over $A$ if and only if its underlying $E_{n-1}$-
algebra is $n$-dualizable.
Read the full abstract ▾
Machine Learning
Scaling Properties of Same-Family On-Policy Distillation
Yuntai Bao, Qinfeng Li, Guoqing Jiang et al. (+5 more)
*Reinforcement learning (RL)* can induce substantial
reasoning capabilities in large
language models (LLMs), but how much of this capability transfers across model scales, and how quickly, remains unclear. We study the scaling properties of *on-policy
distillation (OPD)* across *weak-to-strong*, *same-base*, and *strong-to-weak* teacher–student setups. We find that early OPD training dynamics uniformly exhibit a regular *useful-transfer* regime, in which held-out accuracy (the *gold score*, $G$) rises approximately linearly in $d=\sqrt{\mathrm{KL}(\pi_\theta \Vert \pi_{\mathrm{ref}})}$, the square root of
token-level reverse KL divergence from the student initialization. In every observed weak-to-strong pair, the student's peak gold score exceeds its teacher's own, so a compact RL expert can transfer capability to a much larger student via OPD. To estimate OPD outcomes, we fit *power laws* for how $G_{\mathrm{peak}}$ and the slope of the useful-transfer regime scale with student and teacher parameter counts and with teacher gold score. These laws show that peak gold score improves with teacher scale only up to roughly the student's scale, and that at a matched gold score smaller teachers transfer better, so a teacher's score alone does not define its supervision value. We also study the scaling effects of two OPD variants, bootstrapping weak-to-strong OPD, and the degree of on-policy supervision.
Read the full abstract ▾
All authors Yuntai Bao, Qinfeng Li, Guoqing Jiang, Liwei Chen, Zhiheng Qin, Xuanping Li, Wenqi Zhang, Xuhong Zhang
Combinatorics
Completing shellings with $d-2$ extra vertices
SuHo Oh
Simon's
conjecture (1994) asserts that any pure $d$-dimensional shellable complex on $n$ vertices can be extended to the $d$-skeleton of the simplex on $n$ vertices, one facet at a time, while maintaining shellability. Bolognini and Sentinelli (2026) recently disproved it for every $d \geq 3$. We show that the
conjecture becomes true once $d-2$ new vertices are allowed, and that the same holds with $k$-decomposability in place of shellability for any $k \geq 1$. We also show that the number $d-2$ cannot be lowered. Inflating the counterexample of Bolognini and Sentinelli gives, for every $d \geq 3$, a $1$-decomposable complex that cannot be extended in this way with only $d-3$ new vertices, even while merely maintaining shellability.
Read the full abstract ▾
Machine Learning
BiasReducer: Adaptive Bias Mitigation for Reward Models
Shuang Liu, Yongliang Miao, Yanguang Liu et al. (+2 more)
Reward models score responses from large
language models (LLMs) and guide LLM training toward human preferences. However, reward models can favor superficial attributes such as length or confidence, leading LLMs to produce higher-scoring but not more correct responses. Existing mitigation methods either retrain the reward model or apply a fixed correction to one known bias, such as a preference for longer responses. Retraining requires additional data and computational resources, while existing editing methods require the target bias to be specified in advance and use a fixed edit for that bias. To this end, we propose BiasReducer, a lightweight framework that edits only the linear reward head and selects the relevant edits for each new
dataset. First, BiasReducer uses a sparse autoencoder (SAE)-style encoder to learn which attributes (e.g., length and confidence) the reward model is sensitive to. Second, it learns how to reduce the reward model's dependence on each attribute by determining which direction to adjust the reward head and how much to adjust it. Third, for a new
dataset, it ranks the attributes by their influence on reward scores, selects the relevant ones, and edits the reward model accordingly. BiasReducer consistently improves reward-model robustness to biases toward superficial response attributes. Across five reward models, BiasReducer-M improves the three
benchmarks by 8.3, 18.0, and 6.9 percentage points on average, outperforming the two training-based baselines. The gains transfer downstream, reducing unnecessary verbosity and sycophancy while maintaining comparable judged quality.
Read the full abstract ▾
All authors Shuang Liu, Yongliang Miao, Yanguang Liu, Haoyi Xiong, Mengnan Du
Analysis of PDEs
Blow-Up of Modified Navier–Stokes Equations in Nonhomogeneous Fourier Spaces
Jamel Benameur and Lotfi Jlali
We study the three-dimensional incompressible Navier–Stokes equations with a componentwise nonlinear damping term of the form $\alpha\sum_{k=1}^3 u_k^{2m+1}e_k$ in
Fourier and
Fourier– Gevrey spaces based on $\mathcal{X}^0(\mathbb{R}^3)$. We first establish local-in-time existence and uniqueness of solutions and obtain the corresponding unique maximal solutions. We then derive blow-up criteria in the
Fourier–Gevrey framework, including an integral blow-up criterion in a weaker exponential weight. Moreover, we establish a quantitative lower bound for the solution near a possible finite maximal existence time. By iterating the loss of exponential weight, we finally obtain a corresponding lower bound in the unweighted
Fourier space $\mathcal{X}^0(\mathbb{R}^3)$.
Read the full abstract ▾
Statistics Theory
A Second-Order Extension of Hájek's Convolution Theorem with Statistical Applications
Junichi Hirukawa, Masanobu Taniguchi and Marc Hallin
For a class of regular
estimators, Hájek, in his celebrated “Convolution
Theorem,” showed that the asymptotic distribution of a regular
estimator is the convolution of the distribution of an efficient
estimator and some residual distribution. This result constitutes the foundation of the concept of asymptotic efficiency of regular
estimators. In this paper, we provide a second-order version of that classical result. Introducing a class of second-order regular
estimators with a valid Edgeworth expansion, we derive their asymptotic distribution under contiguous alternatives and show that it is the convolution of the second-order efficient distribution and some second-order residual distribution. This constitutes a second-order extension of Hájek's convolution
theorem. Based on this, we introduce a concept of second-order robustness for second-order regular
estimators. For a class of general Bayes
estimators and minimum contrast
estimators in time series models, this second-order robustness is used (i) in the characterization of second-order robust priors, (ii) in a comparison between the second-order robustness of maximum likelihood and Whittle
estimators.
Read the full abstract ▾
Computation and Language
LLM Alignment–Utility Asymmetry under Semantic-Preserving Transformations
Mohan Li, Chengyu Yu, Francesco Sovrano et al. (+2 more)
Large
Language Model (LLM)
alignment is intended to ensure that models remain helpful and safe, but its stability under input distributional shift is not yet fully understood. Prior work shows that aligned models can fail under jailbreak
prompts, alternative encodings, and cross-lingual transfer, yet these failures are usually studied as attacks rather than controlled probes of
alignment generalization. Moreover, existing evidence is largely grounded in natural
language variation already represented during pretraining, leaving unresolved whether
alignment generalizes with semantic content or remains tied to superficial surface patterns. In this paper, we study this question using synthetic semantic-preserving transformations that are rule-based and invertible, preserving task-relevant meaning while shifting inputs beyond standard linguistic variation. Across four open-weight and four commercial models, under both fine-tuning and in-context learning, we use these transformations as a probe of
alignment generalization and identify an empirical pattern we term
Alignment–utility asymmetry: once models can operate effectively on transformed inputs, task utility is often substantially retained while
alignment failure increases more sharply. For example, adapted GPT-4.1 mini shows only limited utility degradation under transformation while its harmful rate rises from 13.3 to 74.3; Gemini 3 Flash similarly retains near-original utility while its harmful rate increases from 2.3 to 43.0. Taken together, these results suggest that semantic-preserving distribution shifts can expose a recurring gap in how utility and
alignment generalize in current LLMs.
Read the full abstract ▾
All authors Mohan Li, Chengyu Yu, Francesco Sovrano, Marc Langheinrich, Martin Gjoreski
Computer Vision
Region-Local Copula Evidence Fusion for Heterogeneous Remote Sensing Change Detection
Zhiyuan Ji, Junjun Yin and Jian Yang
Superpixel copula models provide stable regional evidence for heterogeneous remote sensing change detection, but a single label per region limits localization within mixed superpixels. This letter develops a region-local copula evidence fusion method that retains the regional decision structure while introducing spatially varying local dependence anomalies. Independently fitted local models characterize departures from unchanged cross-image relationships. Reference ranking and an upper-tail gate transform these anomalies for fusion with continuous regional confidence. We derive the resulting regiondependent local decision threshold and identify a condition under which gating is equivalent to reparameterizing ungated fusion. On Lake and UK, whole-image optimized configurations achieve kappa coefficients of 0.78136 and 0.90817 and improve mixedregion and boundary decisions. Four-fold retrospective spatial validation over ten training subsets confirms complementary local information, with ungated reference fusion increasing mean kappa by 0.00693 and 0.01793. Fixed gating yields a larger UK gain of 0.03353 but only 0.00041 on Lake. These results support regional-local dependence interaction, while showing that calibration and gating have scene-dependent benefits.
Read the full abstract ▾
Signal Processing
Secure Near-Field ISAC with Pinching-Antenna Systems: Exploiting Constructive-Destructive Interference
Dimitrios Bozanis, Vasilis K. Papanikolaou, Sotiris A. Tegos et al. (+3 more)
This paper investigates secure near-field integrated sensing and communication (
ISAC) with pinching-
antenna systems (PASS), where a base station serves legitimate users while localizing a target that may also eavesdrop on the transmitted data. We jointly design symbol-level precoding (SLP) and the transmit and receive pinching-
antenna positions to minimize the Cartesian position error bound (PEB) averaged over the data symbols. Constructive interference ensures reliable reception at the legitimate users, whereas destructive interference drives the target's observation toward incorrect symbol regions without sacrificing the illumination required for sensing. We derive the PEB for unknown complex target reflectivity and establish an exact factorization into transmit-illumination and receive-geometry terms. This decomposition reveals the distinct sensing roles of the two apertures, enables separate receive-placement and joint transmit-placement/SLP optimization, and provides analytical insights into local position identifiability, the range-information limitations of compact arrays, sensing-oriented receive placement, and in-waveguide attenuation. Building on these results, we develop an efficient
algorithm by combining geometry-informed receive initialization, one-dimensional transmit-position searches, and convex per-symbol precoding. Numerical results demonstrate sub-millimeter localization, erroneous symbol decisions at the target, an expanded feasible operating region, and substantial gains over fixed-placement PASS and fully digital extremely large-scale
MIMO (XL-
MIMO)
benchmarks.
Read the full abstract ▾
All authors Dimitrios Bozanis, Vasilis K. Papanikolaou, Sotiris A. Tegos, Panagiotis D. Diamantoulakis, Christos Masouros, George K. Karagiannidis
Nuclear Theory
Structure and decay of $^{21}$B resonances
R. Alvarez-Rodriguez and E. Garrido
The structure of the
neutron-rich
nucleus $^{21}$B is investigated within a three-body model consisting of a $^{19}$B core and two
neutrons. The three-body wave functions are computed using the hyperspherical adiabatic expansion method combined with the complex-scaling technique. Four potential parametrizations compatible with the available experimental information on $^{20}$B are considered. The calculations predict a low-lying $\frac{3}{2}^-$ state below the experimentally observed resonance at 2.47~MeV. An analysis of the dependence on the three-body interaction shows that this state cannot be shifted to the experimental resonance energy; instead, it remains either bound or becomes a near-threshold resonance with a rapidly increasing width. These results indicate that the observed resonance is more naturally interpreted as the second $\frac{3}{2}^-$ state, whereas a lower-lying $\frac{3}{2}^-$ state, dominated by an $s_{1/2}$ configuration, has so far eluded experimental observation. Momentum correlations are obtained for the experimentally observed resonance. In addition, the existence of other resonances is investigated.
Read the full abstract ▾
Probability
Concentration of empirical entropy and transfer entropy for non-regular chains with unbounded memory
José O. Santana and Ricardo F. Ferreira
We study plug-in
estimators of the
entropy rate and the transfer
entropy rate for stationary finite-alphabet chains with unbounded
memory. Under a coalescent coupling-from-the-past assumption, we establish non-asymptotic concentration inequalities around the expectations of the
estimators. Under additional uniform non-nullness and exponential beta-mixing assumptions, we control the corresponding biases and obtain concentration around the
entropy and transfer
entropy rates, together with almost-sure consistency. The results apply to chains governed by g-functions that need not be globally continuous.
Read the full abstract ▾
Artificial Intelligence
MassAlloc Attention: Let Attention Allocate Its Own Compute
Jingze Shi, Zhangyang Peng, Xianduo Li et al. (+6 more)
FullAttn often assigns negligible normalized mass to much of the causal score space, yet dense kernels execute the complete post-score path after forming each QK tile. We introduce MALA, a fused
attention primitive that preserves score access to every legal causal interaction and uses normalized contribution to allocate post-score computation. Forward uses its evolving online-softmax normalizer, while backward reuses the finalized normalizer to derive nested retained support using only standard
attention state. A common tolerance governs training and inference, allowing for adaptive retention of the work. MALA reduces low-contribution post-score computation. A matched-work study at 8K isolates the benefit of distribution-adaptive allocation: under exactly matched total post-score work, MALA approaches a per-instance reference-mass oracle, with mean omitted mass of 0.0188% versus 0.0182%. Across context lengths from 1K to 32K
tokens, the same tolerance maintains low output and gradient errors relative to the reference. Across a broader controlled associative-recall comparison, MALA closely tracks FullAttn as context grows, reaching 89.67% accuracy at 8K compared with 89.97% for FullAttn. In an
attention-operator
benchmark at 128K
tokens with tensor parallelism, MALA reduces forward and backward latency during training by 2.2x and 3.0x and decoding latency during inference by 1.6x relative to FullAttn. Across scaling-law training from 0.6B to 14B parameters, MALA closely tracks FullAttn in perplexity while reducing total training FLOPs. The resulting 14B models and 32B models from separate continued training achieve comparable knowledge,
reasoning, and long-context
retrieval scores to FullAttn. These results indicate that allocating post-score computation according to normalized
attention contributions can retain the evaluated capabilities of FullAttn while reducing
attention computation.
Read the full abstract ▾
All authors Jingze Shi, Zhangyang Peng, Xianduo Li, Yanlin Qi, Xiaotian Lin, Haoxian Chen, Liangdong Wang, Guang Liu, Yuyu Luo
Computer Vision
ProDyGS: Dynamic Gaussian Splatting from a Single Static Monocular Camera
Ugo Leone Cavalcanti, Fabio Tosi, Matteo Poggi et al. (+4 more)
We present ProDyGS, a novel dynamic 3D Gaussian Splatting framework for high-quality novel view synthesis from
videos captured by a single static camera. While existing methods rely on multi-view setups or significant camera motion for
geometric constraints, our approach addresses the challenging scenario where multi-view supervision is completely absent. We overcome this limitation by generating synthetic multi-view supervision through depth-guided proxy image synthesis. Specifically, we estimate temporally consistent depth maps using foundational
monocular depth networks, then construct 3D Gaussian representations that generate proxy images from arbitrary viewpoints. A deformation network learns temporal dynamics by warping canonical Gaussians using this augmented supervision. Experiments on the DyNeRF
dataset demonstrate that our method achieves state-of-the-art performance while requiring only
monocular depth estimation as external supervision, outperforming approaches that rely on stronger priors such as scene flow.
Read the full abstract ▾
All authors Ugo Leone Cavalcanti, Fabio Tosi, Matteo Poggi, Andrea Conti, Vladimir Zlokolica, Valerio Cambareri, Stefano Mattoccia
Probability
Free energy of non-convex multi-species spherical spin glasses
Hong-Bin Chen and Jean-Christophe Mourrat
We identify the limit free energy of all mean-field multi-species
spin glasses defined on products of spheres. In particular, we do not make any convexity assumption on the covariance function of the
Hamiltonian, and we allow for the presence of external fields. We also obtain a one-species reduction of the formula in the special case of balanced models.
Read the full abstract ▾
Quantum Physics
Exact Gaussian Entanglement Dynamics and Initial-State Control in Coupled Parametric Oscillators
Roberto Bernal-Jaquez, Hector H. Hernandez-Hernandez, Alexander Schaum et al. (+1 more)
We study two bilinearly coupled harmonic oscillators driven by a common time-dependent spring. A fixed normal-mode transformation reduces the pair to two parametric oscillators, each solved exactly by its Lewis-Riesenfeld invariant and Ermakov-Pinney amplitude, including the ladder-operator construction and the Lewis-Riesenfeld phase. For invariant-
vacuum inputs the physical two-mode covariance
matrix is obtained in closed form, and the logarithmic negativity collapses to $E_{N}=\frac{1}{2} \operatorname{arcosh}(X/2)$, where a single dimensionless invariant $X \geq 2$ measures the relative phase-space deformation of the two normal modes. We give a normalization-explicit account of the Duan–Simon comparison: after local
symplectic optimization the Duan test and the positive partial transpose share the
entanglement threshold for this pure exchange-symmetric Gaussian family, but the optimized Duan quantity is a witness, not an independent measure, with $D_{\rm opt}/D_{\rm sep}=e^{-E_N}$. We carefully separate a genuinely separable product preparation from the already-entangled ground state of the coupled
Hamiltonian, and use only the former for claims of
entanglement generation. Under sinusoidal stiffness modulation a separable input shows bounded correlations away from resonance and sustained growth near parametric resonance, and we demonstrate numerically that the initial width and chirp act as two independent, experimentally accessible controls of the
entanglement at a chosen target time. An $SU(1,1)$ formulation identifies $2 E_{N}$ with the
hyperbolic separation of the two normal-mode squeezing trajectories on the $SU(1,1)/U(1)$ disk and shows that
entanglement is governed by both the relative squeezing magnitude and the relative squeezing angle; ; we exhibit regimes in which nearly equal magnitudes still produce strong
entanglement through the angle alone.
Read the full abstract ▾
All authors Roberto Bernal-Jaquez, Hector H. Hernandez-Hernandez, Alexander Schaum, Guillermo Chacon-Acosta
Information Theory
One-Step Generative Modeling via Unbalanced Optimal Transport
Yirong Shen, Mengfei Xia, Junpeng Jing et al. (+2 more)
Drifting models enable one-step generation by amortizing distribution transport into training, but this efficiency places greater demands on the transport field estimated at each update. In large-scale training, the field is computed from finite mini-batches of generated and real samples, which provide only imperfect approximations to the underlying distributions. Balanced optimal transport enforces exact mass matching within every mini-batch, making the estimated field sensitive to the particular composition of the real-data batch. We find that generated and real samples should be treated asymmetrically: letting the mass assigned to real samples adapt while keeping every generated sample fully transported improves generation across six feature-space metrics in controlled ablations, and is more robust to the relaxation strength than relaxing both marginals simultaneously, which falls below balanced transport under stronger relaxation. Motivated by this observation, we propose Unbalanced Optimal Transport Gradient Flow (UOT-GF), which keeps the generated-sample marginal fixed and relaxes only the real-data marginal. Under identical settings at DiT-B/2 on ImageNet-256, UOT-GF improves Fréchet Inception Distance (FID) from 1.53 to 1.46 over the balanced W-Flow baseline; scaling the same recipe yields 1.34 and 1.22 FID at L/2 and XL/2, the best FID among the one-step models we compare. We further derive the induced UOT transport force, establish a kinetic Vlasov–Fokker–Planck formulation whose overdamped zero-temperature limit recovers the drifting dynamics, and characterize non-target stationary states together with sufficient conditions for convergence.
Read the full abstract ▾
All authors Yirong Shen, Mengfei Xia, Junpeng Jing, Lu Gan, Cong Ling
Materials Science
Site-Selective Yttrium Substitution in Ti3AlC2 Enables Interlayer Engineering and Li Transport in Ti3C2Tx cathodes for High-Power Energy Storage
Tetiana Boichuk, Andrii Boichuk, Mahesh Eledath Changarath et al. (+5 more)
In this work, we introduce a precursor-level strategy by substitutionally doping the Ti3AlC2 MAX precursor with a small amount of Yttrium during top-down MAX preparation. This allows permanent structural changes that are retained in the resulting Ti3C2Tx MXene. Density functional theory (DFT) calculations identify the outer Ti(4f) site, facing the Al layer, as the preferred incorporation site for Y under the Al-excess synthesis conditions used here, rather than the inner Ti(2a) or Al(2b) sites. This site-specific substitution is consistent with the experimentally observed
lattice expansion and carbide-like Y bonding and provides a microscopic basis for the structural changes retained in the resulting Ti3C2Tx MXene. Y incorporation expands the MXene interlayer spacing due to an increase in the amount of hydroxyl surface terminations, creating more favorable pathways for lithium-ion transport. Comprehensive structural characterization (including XRD, XPS and TEM) confirms these modifications. As a result, the Y-doped Ti3C2Tx cathode shows
diffusion-controlled charge storage with a Li
diffusion coefficient of 10-9-10-11 cm2/s. It delivers a reversible capacity of about 130 mAh/g at 0.2 C and retains 74 mAh/g after 1000 cycles at 2 C with nearly 100% Coulombic efficiency. The electrode also achieves a high power density of up to 3970 W/kg, outperforming the undoped MXene and previously reported multilayered MXene cathodes. These results show that substitutional doping of the MAX precursor is a promising strategy for controlling the interlayer structure of MXenes and improving ion transport, providing a new route for the development of high-power energy-storage electrodes.
Read the full abstract ▾
All authors Tetiana Boichuk, Andrii Boichuk, Mahesh Eledath Changarath, Joao Fonseca, Said Agouram, Marie Finas, Alejandro Molina-Sanchez, Juan F. Sanchez-Royo
Cryptography and Security
Learning to Refer: Client-Resolved Generation for Privacy-Aware Language Models
Jeongho Yoon, Chanhee Park, Yongchan Chun et al. (+5 more)
Cloud-based large
language models (LLMs) require users to disclose plaintext data to service providers, creating
privacy risks in sensitive domains. Existing
privacy-preserving approaches often trade utility for protection, incur substantial computational or communication overhead, remain vulnerable to reconstruction from intermediate representations, or protect only a subset of the training and inference pipeline. We introduce Client-Resolved Generation (CRG), a genera- tion interface that separates server-side generation from the lexical realization of input-derived content. The client transmits only pooled and noise-perturbed rep- resentations, while input-derived output content is represented using request-local positional references and resolved to its original strings only on the client. This interface protects private input and input-derived output content during both train- ing and inference while allowing the service provider to keep its proprietary model parameters hidden from the client. At the same time, exact lexical reuse remains possible without directly exposing the reused content on the provider-visible gen- eration path. We evaluate CRG on medical and document-grounded QA, sensi- tive identifier transfer, and tool calling, together with reconstruction and raw-logit leakage analyses. On SealTools, CRG improves complete-call exact match from 57.3% to 79.9% over the input-
privacy framework PPFT, with larger gains as more required output content can be resolved through references. Together, these results show that CRG provides a practical interface for
privacy-sensitive cloud LLMs by reducing plaintext exposure across both input and output pathways while preserv- ing task utility and server-side model confidentiality.
Read the full abstract ▾
All authors Jeongho Yoon, Chanhee Park, Yongchan Chun, Duong Tuan Thanh, Sungbin Han, Chanjun Park, Hyeonseok Moon, Heuiseok Lim