Analysis of PDEs
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Friday 4 September
Let $P$ denote normalized nearest-neighbor averaging on $\mathbb{Z}^d$ and set $A=I-P$. We consider $\partial_tu+A^su=f(u),t>0, x\in\mathbb{Z}^d, 0<s<1.$ For a Fisher-KPP reaction with $a=f'(0)>0$, compactly supported data have logarithmic propagation rate $a/(d+2s)$. In the logistic case, each fixed level set lies between two constant multiples of $\exp(at/(d+2s))$; this localization remains valid for data with the critical algebraic tail. Nondecreasing front-like data in one dimension instead propagate at rate $a/(2s)$, and the resulting acceleration rules out planar traveling fronts of finite speed in every dimension. We also analyze the singular limit $s\uparrow1$. If $\tau_s=-\log(1-s)$, then throughout $at<\tau_s$ compactly supported solutions follow the Wulff shape of the local nearest-neighbor equation. After a constant multiple of $\tau_s$, their propagation is exponential. Half-space data exhibit the corresponding transition from the directional local speed to the exponent $a/(2s)$. The analysis rests on sharp bounds for the subordinated kernel, algebraic barriers, a compactness–Liouville argument, and a decomposition of the fractional walk into its nearest-neighbor and rare long-jump parts.
Algebraic Geometry
Parahoric motivic Hecke categories in equal and mixed characteristic
We study constructible $\mathbb{Z}[\tfrac{1}{p}]$-linear étale motives on parahoric Hecke stacks associated with quasi-split tamely ramified reductive groups. We prove a canonical equivalence between their equal-characteristic and Witt-vector incarnations that is monoidal with respect to convolution. This extends Bando's comparison, which concerns split reductive groups and constructible étale sheaves. The proof combines the study of a family of affine Grassmannians for Pappas–Zhu group schemes with the use of the motivic nearby cycles functor.
Read the full abstract
Quantum Physics
Non-local games and communication complexity with noisy entanglement
We study the impact of noise on the theories of quantum nonlocality and entanglement-assisted communication complexity. We consider non-local games and entanglement-assisted communication complexity in a model where Alice and Bob may share arbitrarily many noisy EPR pairs. We study four noise models: depolarizing noise, unital noise, biased reset noise, and erasure noise. Our results are as follows: 1. We upper bound the value of the CHSH game under all these noise models in terms of the noise parameter, without any assumptions on the measurements used in the strategy. 2. We prove a parallel repetition theorem for general non-local games under all noise models except biased reset noise; we prove an improved parallel repetition theorem for unique games. Our parallel repetition rate is smaller than the quantum parallel repetition rate for CHSH in a nontrivial noise regime. 3. Using our unique-game parallel repetition theorem and a relation defined by the CHSH game, we prove a separation between communication complexity with noisy vs noiseless entanglement. This implies an $\Omega(n)$ two-way communication lower bound for distilling $n$ EPR pairs from noisy EPR pairs, in the same nontrivial noise regime. 4. We show that any interactive entanglement-assisted communication protocol can be simulated by an SMP communication protocol with noisy shared randomness, with an exponential blowup in communication. This generalizes a known result on the simulation of noiseless shared randomness with noisy shared randomness. 5. We show a polynomial lower bound on the number of copies of noisy EPR pairs required to compute the Equality function with constant communication, under all four noise models. The previous result gives a matching upper bound, and moreover, this answers an open question in the literature on whether logarithmically many noisy shared bits suffice for communication.
Read the full abstract
Statistical Mechanics
On the growth of operator entanglement in brickwork circuits with Yang–Baxter gates
We study the operator entanglement of local operators in one-dimensional brickwork circuits whose two-site gate satisfies the braid relation; throughout this work, we call such a gate a Yang–Baxter gate. We establish upper bounds for several structured, overlapping classes of Yang–Baxter gates. We show that the operator Schmidt rank remains uniformly bounded in time for all qubit Yang–Baxter gates and, in arbitrary local dimension, for permutation gates obtained from non-degenerate Yang–Baxter maps. We also show that it grows at most polynomially for involutive dual-unitary Yang–Baxter gates and for arbitrary phase dressings of permutation gates obtained from non-degenerate Yang–Baxter maps. These results imply, respectively, constant and logarithmic upper bounds on the operator entanglement. Conversely, we construct a seven-state involutive Yang–Baxter gate without dual unitarity and a one-site operator whose exact operator Schmidt rank grows exponentially, although the corresponding operator entropies remain undetermined. Entanglement growth in the general Yang–Baxter case remains open. All proofs and selected examples were constructed by ChatGPT.
Read the full abstract
Materials Science
Symmetry-protected triplet Weyl complexes
The Nielsen-Ninomiya theorem dictates that Weyl nodes must appear in pairs of opposite chirality to preserve global charge neutrality. However, in crystals, specific crystalline symmetries can stabilize multi-Weyl nodes, circumventing this pairwise constraint and enabling compensated Weyl complexes with mixed chiral charges. The minimal configuration of this type is a triplet Weyl complex (TWC), comprising exactly three Weyl nodes. Here, we systematically investigate the symmetry conditions required to realize TWCs. By screening all 1651 magnetic space groups (MSGs) in both spinless and spinful systems, we establish that: (i) Only TWCs with charge magnitudes of $\{1,1,2\}$ and $\{1,2,3\}$ are permitted; (ii) the $\{1,1,2\}$ configuration can be realized in 166 spinless MSGs and 70 spinful MSGs; and (iii) the $\{1,2,3\}$-TWCs, which has not been reported before, can occur in 10 MSGs for both spinless and spinful cases. We explicitly demonstrate the existence of $\{1,2,3\}$-TWC in a tight-binding model. Furthermore, we present the first electronic realization of $\{1,1,2\}$-TWC topological semimetal state in the chiral carbon allotrope DZQH-C$_{36}$, in which the three Weyl nodes form a collinear configuration, leading to a characteristic “S”-shaped surface Fermi arc pattern. Our findings uncover novel topological states featuring mixed chiral charges and provide guidance for exploring their physics in concrete material systems.
Read the full abstract
Cryptography and Security
Understanding the Privacy-Preserving Potential of HTTP/2 Against Webpage Fingerprinting
Website fingerprinting (WF) attacks can infer which webpage a user visits from encrypted HTTPS traffic alone, compromising privacy even without decryption. WF defenses commonly shape traffic through noise, padding, delays, or flow splitting, yet they are most often studied from the perspective of encapsulating protocols like Tor or VPN rather than at the application layer (HTTP). In this work, we focus on application-layer defenses enabled by the most widely deployed version of HTTP, HTTP/2. We demonstrate how known defenses can be emulated through HTTP/2 features at the client side (HTTPOS, LLaMA, FRONT, Tamaraw) and the server side (ALPaCA, Tamaraw). We further show that HTTP/2 features, such as proactive resource suggestion, multiplexing, and flow control, offer untapped potential for lightweight yet effective defenses deployable at both endpoints. We evaluate these defenses using a unified blueprint that calibrates defense parameters per dataset, then combines practical attacks, information-theoretic leakage estimates, and overhead measurements. For each defense, this framework identifies the strongest hyperparameter-tuned fingerprinting model and estimates the residual uncertainty induced by the defense using two information-theoretic leakage estimators, all while accounting for the defense's privacy-overhead trade-offs.
Read the full abstract
General Relativity and Quantum Cosmology
Probing Neutron Star Equation of State Universality with Gravitational Waves
Current neutron star equation-of-state inferences assume that a common equation of state describes all neutron stars. However, {poorly understood dense-matter physics may give rise to distinct long-lived stellar subpopulations following different mass–tidal-deformability relations, invalidating this assumption. In this Letter, we demonstrate that next-generation gravitational-wave observatories enable neutron star equation-of-state universality to be transformed from an assumption into a testable hypothesis. By combining hierarchical Bayesian inference with a piecewise-polytropic parametrization of the neutron star equation of state, we establish a framework for testing equation-of-state universality with next-generation gravitational-wave observatories. We find that about 30 (50) loud binary neutron star mergers with network SNRs above 100 are sufficient to provide compelling evidence for departures from equation-of-state universality, even if only 20% (10%) of the population exhibits one of the representative non-universality parametrizations considered here.
Read the full abstract
Cryptography and Security
TIER: Threat Implicitness Benchmark for Evaluating LLM Safety Behaviors
Current LLM safety benchmarks largely rely on binary metrics, overlooking how models respond to harmful prompts with varying threat implicitness. We introduce TIER, a Threat Implicitness Benchmark for behavioral safety evaluation of LLMs. TIER covers four risk domains and four threat levels, from explicit harmful requests to sophisticated jailbreaks. Responses are assessed using a six-label behavior scale and two independent LLM judges. Experiments on six open-weight LLMs show that safety behaviors evolve gradually across threat levels rather than shifting directly from refusal to compliance. Contextual prompts yield the most diverse behaviors, while jailbreaks reveal the largest robustness gaps. Furthermore, models with similar Attack Success Rates can exhibit distinct response distributions, highlighting the need for behavior-aware LLM safety evaluation.
Read the full abstract
Emerging Technologies
Beyond Maintenance Manual Multimodal RAG: Suggesting What Tool
Aircraft technicians are required to consult the maintenance manual (MM) for nearly every task, and locating the relevant procedure across hundreds of pages remains time-consuming. Multimodal retrieval augmented generation (MRAG) has been proposed to address this, allowing technicians to retrieve procedures, together with the accompanying figures, through natural-language queries. However, retrieval alone does not tell the technicians which tools the task requires. The MM identifies special tools only when the corresponding step is reached, and it does not state hand tool requirements at all; to select hand tools, technicians are required to find the hardware dimension from the illustrated parts catalog (IPC) and infer the right tool from it. We therefore propose MRAG-SWAT, an extension of the MRAG pipeline that returns the required hand tools and special tools alongside the retrieved procedure. The framework was implemented for the Lycoming IO-360-N1A engine and demonstrated on eight test queries. By presenting the correct tools together with the procedure, MRAG-SWAT may help reduce repeated trips to the tool crib, prevent damage to aircraft caused by improper tool selection, and thereby avoid additional maintenance tasks and support continued airworthiness.
Read the full abstract
General Economics
Creators Have Difficulty Abandoning Ideas They Generated
Creativity researchers often distinguish between two stages of the creative process: generation versus selection. While much is known about the psychology of idea generation (e.g., the factors that lead to a greater number of novel and useful ideas), less is understood about the nature of selection, or how generation and selection interact. Here we investigate how the act of generating ideas may potentially distort the selection process. Using an incentive-compatible paradigm in which pairs of participants reviewed the same ideas and were rewarded for submitting only high-quality ideas, we find that people submit a greater number of lower-quality ideas when selecting among their own ideas than when selecting among another person's ideas (the Creative Endowment Effect). This effect generalizes across three tasks in two domains and is resistant to an informational intervention (i.e., explicitly telling people about the effect). However, having participants revisit their ideas several months later increases their selectivity. The broader implications for individuals and organizations are discussed.
Read the full abstract
Computer Vision
VoxelFix: Post-Hoc Semantic Correction of Completed 3D Voxel Maps
Semantic 3D maps are increasingly constructed automatically for aerial robotics by integrating learned semantic predictions into 3D representations. While this avoids costly manual 3D annotation, errors in the perception and mapping pipeline can persist in the resulting map, reducing its reliability for downstream autonomous tasks. Existing 3D semantic map refinement methods either rely on the original observations, treat occupancy as part of the prediction problem, or apply non-learned local regularization to completed maps. Instead, we study post-hoc semantic correction, asking whether semantic accuracy can be recovered directly from the completed map while keeping its geometry and occupancy fixed. We introduce \method, a graph-based model that corrects voxel labels based on local geometry and neighboring semantic information. To obtain training pairs, we corrupt contiguous regions of annotated OccuFly maps according to class confusions observed in upstream maps. We evaluate \method on completed OccuFly maps generated from predictions of four independently trained 2D segmentation models. \method consistently improves mIoU by 4.23–5.00 percentage points, with gains broadly distributed across the evaluated semantic classes and particularly strong improvements for tree, roof, and wall. Results on an independently reconstructed out-of-distribution aerial scene further suggest that the learned correction can transfer beyond the environments seen during training.
Read the full abstract
Machine Learning
A Comparative Study of Counterfactual Explainers for Graph Neural Networks Enabling Multiple Types of Graph Edit
Counterfactual explanations for graph-structured data seek to determine minimal and realistic modifications required in an input graph to alter a model's prediction to a predefined output. Although counterfactual explainers that support modifying the graph by both adding and removing edges have recently emerged, there is still a lack of general and efficient methods, especially when considering the quality of the generated explanations. Moreover, the problem remains far from solved, as existing methods exhibit different strengths and weaknesses, often trading off between explanation size, coverage and quality. For this reason, it is important to identify where each method performs well and where it falls short, so as to guide future research in the field. Thus, our study compares six state-of-the-art (SOTA) models on a diverse set of real-world and synthetic datasets, covering both binary and multi-class graph and node classification tasks, and evaluates their performance using diverse quantitative and qualitative metrics.
Read the full abstract
Probability
Phase transitions for percolation of words in one dimension
In this paper, we investigate models of percolation of words on $\mathbb{Z}_+$ with long-range connections. The underlying graph is oriented and constructed according to a sequence of non-negative ranges. In the first model, the sequence of ranges is considered random and, in the second model, deterministic and non-decreasing. In both cases, we establish a phase transition for the occurrence of percolation of all words simultaneously from the origin.
Read the full abstract
Artificial Intelligence
Unifying ICL, SFT, KL-Regularized RL Through a Bayesian Lens
Supervised fine-tuning (SFT), few-shot in-context learning (ICL), KL-regularized RLHF/RLVR, and on-policy distillation are usually treated as distinct post-training paradigms. We develop a unified Bayesian perspective in which each is an instance of a two-step template: construct a (generalized) Bayes or Gibbs posterior from a reference model and a utility signal (log-likelihood, reward, or advantage), then approximate it by a forward-KL projection onto a parametric family, either in-weights (SFT/RL) or in-context (ICL). This yields a single chain of equivalences: few-shot ICL is an amortized projection onto the Bayes posterior predictive, and reward-weighted SFT, reward-weighted ICL, and advantage-weighted SFT are forward-KL projections of reward-induced Gibbs posteriors. The framework explains why supervised warm-up is practically unavoidable for importance-weighted projections, and interprets R1/o1-style reasoning models as combining test-time Bayesian search with training-time amortization. Matched-budget experiments on Qwen3 models corroborate the picture: operators that share their learning-signal granularity produce nearly identical updates when support is good and diverge when it degrades, and reward-weighted projection performs on par with standard baselines.
Read the full abstract
Information Retrieval
Embedding Surgery: Localized Updates for Adaptive Ranking Correction in Dense Retrieval
Dense retrieval systems are core components of modern search engines, recommendation platforms, and retrieval-augmented generation pipelines. They encode documents and queries into dense embeddings, enabling efficient semantic search via vector similarity. However, because document embeddings are computed offline and stored in static indexes, these systems struggle to adapt to user feedback or evolving search intent. To address this limitation, we introduce embedding surgery, a lightweight approach for adaptive ranking correction in dense retrieval. The method applies localized, minimal updates to selected document embeddings at query time, guided by editorial feedback, user interactions, or pseudo-labels from large language models. We formulate embedding surgery as a convex optimization problem that enforces ranking constraints while minimizing modifications to the affected document representations. We integrate embedding surgery into standard dense retrieval pipelines and evaluate it on TREC Deep Learning, TREC Robust, TREC CAsT, and MS MARCO benchmarks. Results show consistent improvements (e.g., up to +60.64% relative improvement in nDCG@10 on DL-Hard under editorial feedback), even under noisy or shifting feedback, with low computational cost and without disrupting the global structure of the embedding space. Extensive experiments show that ranking corrections propagate to semantically related queries and that embedding updates can be applied safely and efficiently to scalable Approximate Nearest Neighbor indexes via simple in-place overwriting, without requiring costly index reconstruction. Finally, embedding surgery complements query adaptation methods such as CoRocchio, yielding additional gains while being more robust to noisy feedback.
Read the full abstract
High Energy Physics (Phenomenology)
The axial-vector nucleon form factor in the meson dominance picture: the role triangle singularities from a dispersive view
The nucleon axial form factor is analyzed in terms of a dispersion relation comprising chiral perturbation theory (ChPT) and perturbative QCD (pQCD) in the low- and high-momentum regimes respectively, which implies a set of superconvergence sum rules for the $A \to N \bar N$ spectral function. In the timelike region below the $N \bar N$ production threshold we include the $a_1$ and $a_1'$ resonances, with finite-width effects modeled through $\rho \pi$ and $\sigma \pi$ intermediate states. Above the $N \bar N$ threshold, we model an infinite tower of radially excited $1^{++}$ resonances with a non-integer power fall-off, eventually matching pQCD. In this setup, we find that both the ChPT and pQCD contributions are tiny. In turn, the dominant $\rho\pi$ triangle provides the leading deviations from axial meson dominance. Our calculation is not a fit; it contains parameters with a priori uncertainties estimates. The axial radius is predicted to be $r_A^2 =(0.28-0.33)~\textrm{fm}^2 = ( 0.47-0.57~\textrm{fm})^2$ in agreement with recent lattice QCD results but inconsistent with the MINER$\nu$A determination. A simple semiempirical formula is provided accounting for the main physical effects.
Read the full abstract
Lattice Field Theory
Equation of state of dense magnetized QCD matter via a robust analytic continuation
We compute the equation of state of dense and magnetized strongly interacting matter using lattice QCD at a non-zero value of the lattice spacing. Our simulations employ $2+1+1$ dynamical flavors of rooted staggered quarks with masses tuned to the physical point. To conform with experimental constraints, we impose strangeness neutrality and isospin asymmetry on our observables. We circumvent the sign problem by simulating imaginary values of the chemical potential and applying the $T'$-expansion scheme to carry out the analytic continuation of our observables to the real chemical potential axis. This novel scheme, adapted to the setup with nonzero magnetic fields, is demonstrated to produce significantly smaller systematic errors in the final observables, compared to the standard analytical continuation at fixed temperature, allowing for better-controlled extrapolations. Our results for the pressure, the entropy density and the interaction measure are tabulated for direct use in heavy-ion phenomenology applications.
Read the full abstract
Analysis of PDEs
Monotonicity formulas in positively curved settings with applications to two-phase free boundary problems
Inspired by the monotonicity formula of Alt, Caffarelli, and Friedman, we consider a natural variant of the ACF functional in positively curved 2-dimensional settings where we integrate over sublevel sets of Green's function rather than disks. We prove sharp almost-monotonicity formulas for our new functional in the case of convex planar domains (both with the pole on the boundary and in the interior) and in the setting of 2-dimensional complete manifolds with Euclidean volume growth and nonnegative Gaussian curvature. The tools include the Schwarz-Christoffel formula and Riesz decomposition using conformal coordinates. As a consequence, using quasiconformal estimates, we give a new Lipschitz bound in the manifold setting for minimizers of the two-phase free boundary problem of Alt, Caffarelli, and Friedman, with the constant depending on only the total integral curvature, in contrast to work of Teixeira and Zhang, which gives constants depending on pointwise bounds for the Riemann curvature tensor and its derivatives. Our methods also recover the Lipschitz bound up to a Neumann boundary of Gemmer, Moon, and Raynor in the planar convex domain case.
Read the full abstract
Differential Geometry
Kähler Einstein manifolds and the Calabi curvature operator
In this paper, we study the Calabi curvature operator on Kähler manifolds. First, we prove that if the Calabi curvature operator on Kähler manifolds satisfies $\frac{{n\left(n + 1\right)}}{2}$-positive (nonnegative), $\frac{{n + 1}}{2}$-positive (nonnegative), and $\left( n-1 \right)$-positive (nonnegative), then the scalar curvature, Ricci curvature, and orthogonal Ricci curvature are positive (nonnegative), respectively. Second, we show that any compact Kähler Einstein manifold satisfying the condition $$\lambda_1+\dots+\lambda_{\alpha}\ge -{\alpha}\theta(n,{\alpha})\bar\lambda,\; {\alpha}\le \frac{n}{2}$$ must have nonnegative constant holomorphic sectional curvature.
Read the full abstract
Earth and Planetary Astrophysics
Modelling Palomar Transients: Constraints from Reflection Geometry and Orbital Altitude
Recent searches of digitised photographic plates from the Palomar Observatory have uncovered tens of thousands of short-lived transients, each visible in only a single exposure. Their morphologies indicate sub-second flashes, consistent with specular reflections from highly reflective objects in near-Earth orbits. In this paper, we present a modelling study using geometric shadow modelling, Monte Carlo simulations, and photometric constraints to infer their physical properties. We examine the observed groupings and alignments, which appear to be consistent with sudden changes in attitude. Assuming a spherical-shell model, the angular profile of the measured transient deficit around the antisolar point is consistent with a population at characteristic altitudes of $\sim 20,000$–25,000 km. A complementary estimate, based on the altitude dependence of the global Earth-shadow deficit, yields a broader characteristic range of $\sim$ 20,000–35,000 km above Earth's surface, extending into the geosynchronous orbital (GSO) region. Under simplified geometric and photometric assumptions and assuming $\sim20{,}000$ to $35{,}786$ km above the Earth's surface, the inferred sizes of the specularly reflecting facets range from centimetre scales up to $\sim3$ m, with characteristic flash durations of $\sim320$ ms and slow rotation rates. We explore both natural and non-natural toy models consistent with the data, and provide constraints to guide future observational searches.
Read the full abstract
Artificial Intelligence
Compact Bellman-Grounded Cognitive Maps for Cost-Aware Navigation
Biological agents navigate familiar environments not by re-solving routes for each new goal, but by reusing a learned map built once and read off as goals change. Existing artificial cognitive-map models mimic this reuse, yet their guidance is not explicitly grounded in additive heterogeneous route costs. Furthermore, they often struggle with memory efficiency: representative state-indexed and high-rank spectral constructions incur substantial storage growth as the environment scales. We present BCM, which grounds a reusable cognitive map in local edge costs through a self-supervised Bellman-grounded objective and a compact coordinate encoding, supporting changing goal queries without per-goal retraining. On weighted grids of up to $N=1600$ nodes, BCM maintains full success and only a 5% mean Gap relative to exact Dijkstra search, compared with about $45\%$ for a connectivity-based spectral baseline. Notably, as the graph size increases from $N=400$ to $N=3600$, its memory footprint grows sublinearly while maintaining competitive performance, making our method scalable to complex environments. Together, these results show that additive route costs can be written into a compact, reusable cognitive-map representation, bridging the gap between biological flexibility and optimal path planning.
Read the full abstract
Differential Geometry
From quadratic integrals to Nijenhuis operators: two-dimensional dictionary
We construct an explicit dictionary between two local classifications in dimension two: normal forms of pseudo-Riemannian metrics with a quadratic integral of the geodesic flow and normal forms of $\operatorname{gl}$-regular Nijenhuis operators. The correspondence is obtained by an explicit change of coordinates constructed from the trace and determinant of the operator.
Read the full abstract
Human-Computer Interaction
Beyond Bias: Participatory and Reflective Approaches to Cultural AI
Generative AI systems increasingly shape cultural production, yet creative intentions, cultural meanings, and interpretive practices often can't be articulated through computational metrics alone. This paper presents Beyond Bias, a collaboration between Gooey.AI and Goethe-Institut India, as a participatory approach to cultural AI which includes collaborative dataset creation, reflective AI tooling, artist-led model fine-tuning, and co-authored governance practices. Across 9 workshops involving over 200 participants, artists and cultural practitioners engaged with AI systems through experimentation, iteration, and collaborative LoRA training. Participants used their AI-generated outputs and visualizations as reflective interfaces for exploring symbolism, memory, authorship, and cultural contexts. Comparing contemporary generative AI outputs with participant fine-tuned outputs helped participants reflect on cultural details missing in big tech AI systems. This paper contributes reflective AI tooling approaches foregrounding transparency, stewardship, and community participation; findings from participatory workshops examining how generative AI visualizations mediate cultural representation and interpretive practice; and a framework for cultural AI grounded in cultural integrity, and reflective practice.
Read the full abstract
Optimization and Control
Duality for Stochastic Control with non-Markovian Random Coefficients
We develop novel duality methods for stochastic optimal control problems under two sources of randomness and non-Markovian random coefficients adapted to just one of them. The Hamilton-Jacobi-Bellman (HJB) equation is a second-order backward stochastic partial differential equation. Our duality theory provides an alternative description of the random value function in terms of a suitable pathwise optimal control problem parameterized by the realizations of one of the Brownian motions. This allows us to regain Markovianity using the theory of rough optimal control problems, leading to rough second order HJB equations that we solve in a suitable viscosity sense.
Read the full abstract
Combinatorics
Layered mixed matrices and reaction networks
The purpose of this work is twofold. In the first part, we consider layered mixed matrices introduced by Murota, relate them to existing notions in combinatorial commutative algebra, and investigate the irreducibility of their determinants. Furthermore, for a layered mixed matrix in combinatorial canonical form, we determine the sparsity structure of its inverse. That is, we characterize which entries of the inverse are nonzero. In the second part, we establish for the first time a formal connection between these algebraic results and the theory of buffering structures for reaction networks developed by Mochizuki and Okada. We identify the lattice of buffering structures with the lattice of order ideals of the block poset of the combinatorial canonical form of the associated layered mixed matrix. This allows us to characterize the reducibility of the symbolic Jacobian determinant as a polynomial in the reaction-rate derivatives, as well as the nonzero sensitivity responses of species concentrations to reaction-rate perturbations.
Read the full abstract
Computation and Language
Improving Language Identification for Code-Switched Utterances with Integer Linear Programming
Automatic identification of code-switched (CS) utterances remains a challenge for language identification (LID) systems, causing such texts to be underrepresented in the training data of Large Language Models. In this paper, we revisit MaskLID, a state-of-the art approach for CS identification, which requires no training and detects arbitrary language combinations. We make three main contributions: (a) we reveal, and address, a major issue of MaskLID: its overreliance on word-level language association scores; (b) we reformulate the underlying optimization algorithm as an Integer Linear Program, enabling us to experiment with a large set of clear and interpretable constraints; (c) each of these improvements vastly improves the baseline system, as we illustrate in experiments involving 10~diverse languages, where we observe a strong boost in performance on CS benchmarks. We release our code and data for reproducibility.
Read the full abstract
Machine Learning
NEAT-POCKET: Pocket-Conditioned Autoregressive 3D Molecular Generation with a Neighborhood-Guided Set Transformer
AI-driven de novo molecular design offers a promising route to accelerate early-stage drug discovery by generating novel ligands directly within target protein binding pockets. We present NEAT-POCKET, a pocket-conditioned extension of the autoregressive NEAT model for 3D molecular generation. NEAT-POCKET generates molecules atom by atom in protein pocket environments while preserving atom permutation invariance and explicitly modeling hydrogen atoms. Benchmarks on the CrossDocked and SPINDR datasets show that NEAT-POCKET achieves competitive structure-based generation performance while sampling substantially faster than existing baselines. Beyond full-molecule generation, NEAT-POCKET naturally enables pocket-conditioned fragment completion, a task directly relevant to lead optimization and scaffold elaboration. These results position NEAT-POCKET as a fast, flexible, and practical framework for structure-based drug design.
Read the full abstract
Statistical Methodology
Post-Corrected Raw-Score Martingale Posterior Sampling for von Mises-Fisher Models
We develop a finite-horizon calibration method for raw-score martingale posteriors, with von Mises–Fisher models as the main worked example. Starting from the maximum likelihood estimator, predictive paths are generated by simulating future observations from the current fitted model and updating the natural parameter by unpreconditioned score increments. The main methodological step is to separate predictive simulation from covariance calibration. Raw-score increments have Fisher-information covariance, whereas Bernstein–von Mises calibration requires inverse-information covariance. We therefore apply a terminal linear correction based on a local information estimate. For more efficient implementation, we also introduce a hybrid version that replaces the omitted tail of the infinite predictive continuation by a Gaussian approximation with matching leading-order quadratic variation. We prove fixed-n convergence, a finite-horizon approximation bound for the Gaussian tail, and a Bernstein–von Mises limit for the hybrid post-corrected sampler under local regularity and consistent terminal calibration. Simulations show that tail correction reduces truncation-induced underdispersion, and an OSCAR ocean-current example illustrates local directional uncertainty summaries.
Read the full abstract
Databases
CAT-LDP: Cloud-edge Adaptive Taxonomy under Local Differential Privacy
Recommender systems are widely used in daily life, but their direct collection and use of user preference data can also lead to privacy leakage. Existing privacy-preserving recommendation methods often find it hard to balance user privacy and recommendation performance. This problem is more serious in implicit-feedback settings, where data sparsity further increases the loss of useful signals caused by privacy perturbation. To solve this problem, we propose CAT-LDP, a cloud-local collaborative recommendation framework under local differential privacy constraints. CAT-LDP combines a hierarchical taxonomy tree with an adaptive privacy budget allocation strategy to keep more useful signals in users' active categories while protecting user privacy. Specifically, users upload perturbed category profiles that satisfy LDP. Based on these profiles, the cloud performs coarse-grained candidate generation, and the local device then carries out fine-grained reranking by using unperturbed local history. Experiments on the Amazon Video Games dataset show that CAT-LDP consistently outperforms its fixed-budget ablation variant and representative baselines on HR@K and NDCG@K under different privacy budgets. The results show that combining category-space modeling with cloud-local task decoupling can effectively reduce noise amplification in long-tail sparse settings and provide a better balance between privacy and utility for implicit-feedback recommendation.
Read the full abstract
Artificial Intelligence
ProCA: Progressive Contrastive Alignment for Robust EEG Visual Decoding
Electroencephalogram (EEG) visual decoding aims to recover visual semantics from non-invasive neural time-series signals, for which robust alignment between noisy neural responses and stable semantic representations is key to achieving high-performance decoding. Despite recent advances in contrastive learning, robust EEG decoding remains challenging because existing methods rely on fixed visual or textual anchors whose semantic relations may become misaligned with EEG representations that vary across trials, subjects, and learning stages. Our empirical evidence shows that this instability appears across both standard EEG decoding protocols and more challenging robustness settings, including strict cross-subject transfer and realistic personalized continual adaptation. We provide a formal analysis showing that fixed semantic supervision can bias optimization when EEG-specific relations evolve, and that structure-agnostic perturbations may distort semantically important EEG components. To address these issues, we propose Progressive Contrastive Alignment (ProCA), a unified and model-agnostic framework for adaptive neural-semantic alignment. ProCA progressively refines class-level contrastive supervision from frozen vision-language priors to EEG-aware semantic relations, and introduces structure-consistent interpolation to constrain feature mixing according to channel-wise and temporal importance. Across subject-dependent, subject-independent, strict cross-subject transfer, and continual adaptation settings, ProCA achieves average relative Top-1/Top-5 gains of 7.4%/3.9%, 10.0%/4.6%, 28.1%/17.8%, and 16.8%/11.6%, respectively.