Cryptography and Security
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Tuesday 1 September
Extended Reality (XR) provides immersive, interactive 3D experiences. To enable these experiences, the devices must track user motion so the system can respond to actions such as grabbing, looking at, or moving an object. However, motion tracking has raised privacy concerns since it records a person's motion patterns. These motion patterns have been studied extensively across various fields (i.e., gait identification and profiling) and have been shown to reveal sensitive information. With the adoption of XR, these patterns became easier to record and obtain than ever. This creates a fundamental privacy tension: motion tracking enables core XR functionality yet requires users to compromise their privacy. Prior systematization-of-knowledge (SoK) studies on XR privacy have examined the field broadly, with motion-related research distributed across several privacy domains rather than treated as a distinct area of study. However, XR motion privacy has gained significant momentum since the prior SoK, with the literature nearly quadrupling in size and thereby warranting a dedicated systematization of this topic. This SoK examines 134 relevant papers on privacy concerns in motion patterns recorded by XR headsets, including how adversaries can obtain users' motion patterns, the inferences they can draw from them, and methods for protecting users. Based on this review, we synthesize a taxonomy of motion modalities, representations, and inference risks; develop an XR motion threat model; systematize the attack and defense approaches in the XR motion literature; identify gaps in the literature; and provide guidelines for future studies evaluating motion privacy mechanisms. Together, our SoK clarifies the state of XR motion privacy and provides recommendations for future evaluations.
Data Structures and Algorithms
Prediction-Assisted Pricing and Admission for LLM APIs with Stochastic Token Consumption
An LLM application often sells or internally allocates several service products: a small or premium model, a short or long token cap, and possibly multiple posted prices. The operational decision is not merely which model answers a prompt. A price changes purchase probability, a token cap changes both user value and the tail of resource consumption, and accepted requests compete for shared compute and premium-model capacity. Demand and output length are initially uncertain, while an offline model may provide useful but imperfect predictions. We formulate sequential pricing and admission with stochastic resource consumption. Each arriving request belongs to an observable segment. The platform chooses a product–price pair or makes no offer; purchase, revenue, and resource use are then random. An offline predictor supplies a uniform, validated error radius for every segment–product cell. We propose Prediction-Clipped UCB (PCUCB), which intersects the offline prediction interval with an online confidence interval, evaluates products using resource shadow prices, and reserves a sample-path envelope before commitment. The prior gives a fast start when accurate, while online learning protects the platform when predictions are coarse. The analysis is modular. On a simultaneous confidence event, regret against a buffered fluid benchmark is bounded by a pacing term plus the cumulative diameter of the intersected intervals. For $J$ segment-product cells and prediction radius $\varepsilon$, this yields \[ \widetilde O\left( \sqrt{T}+(1+\bar\Lambda) \min\{T\varepsilon,\sqrt{JT}\} \right), \] where $\bar\Lambda$ bounds operational shadow prices. Thus the algorithm smoothly interpolates between an almost full-information regime and learning from scratch. Hard feasibility holds on every sample path through reservation envelopes.
Read the full abstract
Computer Vision
Controllable Image Captioning with Prompt-Conditioned Scene Rewards
Large Vision-Language Models produce fluent image descriptions but offer limited semantic control: users cannot reliably specify whether captions should emphasize attributes, relations, or particular image regions. We present Fine-grained Captioning Control Using Scene Rewards (FoCUS), a controllable image captioning method that lets users steer captions toward specific semantic emphases through natural-language control prompts. The core idea is a prompt-conditioned control objective based on scene-graph-aligned component scores. Generated captions are parsed and aligned to scene-graph components such as objects, attributes, and relations. These components are differentially weighted, including negative weights, according to the requested emphasis. We optimize this objective with GRPO and further improve its reliability through a stricter object validity threshold and reasoning-based verification for attribute and relation scoring. To evaluate controllability, we introduce Semantic Control and Precision Evaluation (SCoPE), a benchmark with contrastive Include/Avoid constraints for measuring both target content coverage and out-of-scope suppression. Experiments on two VLM backbones show that FoCUS consistently improves controllability and fine-grained caption quality without degrading general caption performance.
Read the full abstract
Cryptography and Security
Differentially Private Paired Table-Image Multimodal Synthesis
Differentially private (DP) synthesis has been extensively studied for tabular and image data separately, yet many real-world datasets contain images paired with multivariate tabular records. Synthesizing such data is particularly challenging under DP, as the two modalities favor different private learning mechanisms while their dependence must also be preserved. To address this challenge, we propose DP-TabImage, a modality-specialized framework for private paired synthesis. DP-TabImage instantiates the factorization $p(x,y)=p_T(y)p_I(x\;|\;y)$ using a private Probabilistic Graphical Model for the multivariate table distribution and a table-conditioned diffusion model trained with DP-SGD for the conditional image distribution. To facilitate conditional learning under clipped and noisy gradients, we further pretrain the model on private table-image prototypes, pairing privately constructed attribute-conditioned images with tabular vectors derived from the already private tabular model at no additional privacy cost. Experiments on three real-world datasets show that DP-TabImage achieves a strong balance among tabular fidelity, image fidelity, and cross-modal alignment. Our analysis further reveals that visual warm-up primarily improves marginal image fidelity, whereas aligned table-image warm-up is critical for improving cross-modal correspondence. Our source code is available in the GitHub repository, https://github.com/KaiChen9909/TabImage_Syn.
Read the full abstract
Commutative Algebra
Trace ideals of syzygies
We study the behavior of trace ideals under taking syzygies. In particular, when $R$ is numerical semigroup ring and $I$ is a homogeneous ideal in $R$, we obtain an upper estimate for $\operatorname{tr}_R(\Omega^1_R(I))$, and we show this estimate is sharp when $I$ is the conductor ideal $\mathfrak{c}_R$ of $R$. Using this result, we characterize the numerical semigroup rings for which $\operatorname{tr}_R(M) \subseteq \operatorname{tr}_R(\Omega^1_R(M))$ holds for every finitely generated $R$-module $M$. In a similar vein, we characterize numerical semigroup rings for which $\operatorname{tr}_R(\Omega^1_R(\mathfrak{c}_R))=\mathfrak{c}_R$.
Read the full abstract
Computation and Language
A Certificate-Producing Cascade for Equational Implication: The SAIR EQT2 Stage 2 Solver
The SAIR Mathematics Distillation Challenge on Equational Theories asks a solver to classify whether one magma identity implies another and, for either verdict, to return a certificate accepted by a deterministic Lean judge. We present a single-file solver organized as a cheapest-first cascade. Its false branch combines coefficient tests over structured algebra families, bounded finite-model search, an explicit central-groupoid witness, and several infinite-carrier witnesses. Its true branch is a proof-producing ordered unit superposition procedure with Knuth-Bendix ordering, bidirectional demodulation, indexing, memoised substitution, and anytime size deepening. Search results remain outside the trusted base: successful derivations are replayed as small Lean terms, and countermodels are rechecked by the competition judge. The frozen solver is a 189,504-byte Python file with SHA-256 f2392533c9f4c03b.... In local runs through official judge revision 2848228, it produced accepted certificates for all 1,889 rows of the six public sets with no language-model calls. Separate measurements recorded full agreement on the 800 published Stage 1 evaluation-distribution problems, 100 accepted rows in the canonical Marathon manifest without tokens, and 200 accepted rows in the hosted playground. These are regression and playground measurements, not a leaderboard result and not evidence about a hidden set. All quantitative claims are tied to immutable result ledgers; the paper makes no completeness or comparative-superiority claim.
Read the full abstract
Cryptography and Security
PhantomCall: Evading ML Malware Detectors via Function Call Graph Perturbation
Prior adversarial attacks on Windows PE malware detectors target raw bytes, PE headers, or intra-function control-flow graphs, leaving the function call graph (FCG) unexplored as an attack surface. Yet the FCG structure is an important feature in graph-based malware detectors. We present Phan- tomCall, a black-box attack that perturbs the FCG of Windows PE malware by injecting fully executable dummy functions at targeted call sites, adding new nodes and edges to both the CFG and FCG while preserving program semantics. We pair this structural perturbation with classifier-guided search and tunable injection parameters, effective across three archi- tecturally distinct classifiers. Evaluated on a 2025-collected Windows malware corpus against MalConv (raw-byte CNN), MalGraph (graph-based GNN), and SAFE+GNN (pure FCG GNN trained from scratch on a 2024 corpus) at two FPR thresholds, the best PhantomCall variant achieves 85-100% attack success rate across all configurations, exceeding prior state-of-the-art by up to 14.78 percentage points on MalGraph and 95.5 percentage points on SAFE+GNN, and generating evasive variants up to 2.9x faster on average across all targets. For MalConv and MalGraph, the majority of evasions require only a single call site modification, and 86-97% of evaluated evasive variants preserve the original malicious behavior in sandbox-based semantic testing across all configurations.
Read the full abstract
Computer Vision
FTU-Seek: Foundation Model-Guided Hard-Negative Learning for Sparse Functional Tissue Unit Segmentation
Functional tissue units (FTUs), including tertiary lymphoid structures (TLSs), blood vessels, and glands, encode localized immune, vascular, and epithelial organization in histopathology. Accurate quantification of these structures is important for studying tissue architecture and disease-associated tissue organization. However, FTUs are frequently sparse, heterogeneous, and surrounded by large amounts of morphologically similar background tissue, making automated segmentation in whole-slide images (WSIs) challenging. We therefore developed FTU-Seek, a pathology foundation model-guided framework that treats morphology-aware negative-patch selection as a key component of sparse FTU segmentation. FTU-Seek uses frozen multi-depth features from the UNI pathology foundation model to train a patch-level classifier that distinguishes FTU-containing from FTU-absent tissue. Target-absent patches are subsequently ranked according to their predicted target-containing probabilities, and the highest-scoring hard negatives are selected through a static Top$K$ strategy to construct compact segmentation training sets. The framework was evaluated using five-fold cross-validation and internal test cohorts across TLS, blood-vessel, and gland segmentation tasks, with an additional independent 30-WSI held-out cohort for TLS. Positive-only, all-tissue, random-negative, and matched random Top$K$ sampling strategies served as comparators. Segmentation-derived phenotypes were further explored in external TCGA cohorts.
Read the full abstract
Analysis of PDEs
Ellipsoidal Positivity Sets for Fractional Obstacle Problems with Quadratic Forcing
Let \(n\ge1\), \(0<s<1\), \(c>0\), and let \(A\) be a positive definite symmetric matrix. We prove that the unique decaying viscosity solution of \[ \min\{u,\,(-\Delta)^s u-(c-\langle Ax,x\rangle)\}=0 \qquad\text{in }\mathbb R^n \] has the form \[ u(x)=K\max\{1-\langle Bx,x\rangle,0\}^{1+s} \] for some \(K>0\) and some positive definite symmetric matrix \(B\). In particular, its positivity set is an ellipsoid. For \(s=1/2\) and \(n\ge2\), this result, combined with the classification of Fernández-Real and Yu, implies that cubic global thin obstacle solutions with nonempty bounded positivity set have ellipsoidal positivity sets. This proves the conjecture of Fernández-Real and Yu in the nonempty bounded-positivity case.
Read the full abstract
Functional Analysis
On the Set-Valued Katugampola Fractional Integral: Properties and Regular Selections
In this paper, we develop a theory of generalized fractional integration for set-valued mappings for the Katugampola fractional integral, which unifies the classical Riemann-Liouville and Hadamard fractional integrals. The Katugampola fractional integral of a set valued mapping is studied using integrable selections. We study its properties with respect to the Hausdorff metric on the space of nonempty compact subsets of R and several fundamental analytical characteristics including convexity, boundedness and continuity are preserved under Katugampola fractional integration. Furthermore, we establish that both bounded variation and Lipschitz regularity of a set-valued mapping are preserved under its Katugampola fractional integral. We investigate the existence of regular selections associated with the Katugampola fractional integral and show that whenever the original set valued mapping admits a selection with a specified regularity property, the corresponding Katugampola fractional integral does as well.
Read the full abstract
Emerging Technologies
FALCON: Fault-Tolerant Magnetic Tunnel Junction-Based In-Memory Stochastic Architecture for Reliability-Critical Edge AI Applications
As modern data-centric applications such as neural inference and sensor-edge analytics expand, they increasingly encounter the von Neumann memory wall, suffering from excessive data movement overhead and stringent energy constraints. In-Memory Computing (IMC) utilizing emerging non-volatile technologies, such as Magnetic Tunnel Junctions (MTJs), promises to mitigate these bottlenecks. However, conventional binary radix-based IMC architectures suffer from excessive vulnerability to process-induced variations, restricted operating margins, and thermal noise. To bridge the gap between energy efficiency and computational reliability, this work proposes FALCON, a fault-tolerant, MTJ-based in-memory arithmetic architecture integrated with Stochastic Computing (SC). By encoding numerical values into uniform bit-streams, SC naturally absorbs localized soft errors and enables the execution of an essential suite of arithmetic operations using highly compact logic primitives directly within the memory arrays. FALCON integrates a deterministic bit mapping mechanism with reconfigurable logic-in-memory (LIM) structures, eliminating the need to transfer data to external processors or area- and power-hungry random number generators. Experimental results using 14 nm FinFET technology validate the correct functionality of FALCON even under aggressive voltage scaling, severe process variation, and noise injection levels up to 30%, making it a robust framework for reliability-critical edge AI applications. We investigate the proper functionality of FALCON on morphological closing as a realistic noise-tolerant image processing case study.
Read the full abstract
Artificial Intelligence
Value Over Language Model: Detecting Original Contribution in Writing
LLMs have been rapidly adopted across writing tasks, prompting the development of tools for detecting LLM-generated text. Yet, these tools largely measure how much of a document's surface text was written by an LLM and aren't fundamentally designed to measure how much of the information content or ideas originated from the LLM itself rather than being supplied by the user in the prompt. In this work, we design a framework that measures how much value a person adds on top of what a language model could have easily produced by itself. The method requires no training or labeled data and never scores the document's surface text, insulating it from stylistic confounders. Instead, it extracts the document's content at increasing levels of granularity, uses an LLM to reconstruct the document from each partial representation, and compares these reconstructions with those produced from the task description alone. We call this framework Value Over Language Model (VOLM), which measures a document's contribution relative to a replacement-level document that an LLM could produce from the task description alone. We evaluate VOLM with a specific instantiation of this framework across three domains: news articles, ICLR peer reviews, and argumentative essays. VOLM separates human-authored documents from matched LLM-generated documents produced from generic task descriptions, while remaining substantially invariant to content-preserving transformations, including LLM-based reconstruction and round-trip translation. We further find that increasingly constrained content extractors reduce residual differences between LLM-generated and humanized text, demonstrating the importance of disentangling informational content from stylistic variation. We hope these results encourage further work on specialized instantiations of the framework and on assessing human contributions in LLM-assisted writing more generally.
Read the full abstract
Machine Learning
Patterning in Practice: Debiasing Reward Models with Susceptibilities
Reward models trained on human preferences are known to suffer from length, formatting, and other stylistic biases. In this paper we use patterning, which reweights each preference pair according to its measured effect on posterior expectation values of benchmark losses (its susceptibility), to debias a Gemma 2 9B Instruct reward model trained on Skywork-Reward-Preference v0.2. We obtain $+14.2 \pm 1.2$ pp on RM-Bench Hard, the split where style cues point against correctness (mean $\pm$ s.e.\ over 5 seeds), with overall RM-Bench accuracy preserved, comparable to the strongest Hard-split gain reported by the closest published comparator (SteerRM, $+13.2$ pp). We demonstrate in a simple case that the reweighting is interpretable by tracing a side effect of the intervention (a regression on a safety subset of RM-Bench) to a small class of training pairs, which we confirm by ablation. The weights also transfer: those computed on Gemma 2 9B debias Gemma 2 2B and 27B with no recomputation, and transfer partially to Llama 3.1 8B. This is the first application of patterning, a program grounded in singular learning theory, beyond small models and synthetic tasks.
Read the full abstract
Analysis of PDEs
Kinetic Wedge Layers and Diffusive Limit of Neutron Transport in Polygonal Domains
We study the diffusive limit of the steady neutron transport equation with velocity-dependent in-flow boundary data in polygonal domains. On a bounded convex polygon, we construct an approximation by the interior solution, flat side layers, and wedge layers, and prove an algebraic error estimate in $L^\infty$. We also give a complete formulation and well-posedness theory for the kinetic wedge layer, including its algebraic decay. The proof combines a characteristic stability estimate, a weighted Mellin mapping theorem, a two-depth construction and matching scheme, and a shifted superharmonic barrier for the wedge corrector. We also prove the $L^2$ diffusive limit at rate $O(\varepsilon^{1/2})$ on general bounded simple polygons, including those with reentrant vertices.
Read the full abstract
Strongly Correlated Electrons
Correlation-Driven Nonlinear Magnetoelectric Response in an Altermagnet: A Dynamical Mean-Field Study
We investigate the optical nonlinear magnetoelectric effect (NMEE) in a strongly correlated altermagnet using dynamical mean-field theory. Unlike effective band descriptions with an imposed spin splitting, our approach determines the altermagnetic order, electronic spectrum, and optical nonlinear response self-consistently. We find that the NMEE is finite in the altermagnetic phase and vanishes in the paramagnetic phase. Its frequency dependence reflects the spin-resolved spectral structure and provides an estimate of the characteristic altermagnetic spin-splitting scale. Interaction and temperature tuning produce qualitatively different behavior: at low temperature, reducing the interaction strength toward the interaction-driven magnetic phase boundary enhances the response, whereas increasing the temperature suppresses it and drives it to zero above the critical temperature. These results establish the optical NMEE as a probe of correlated altermagnetic order and suggest that tuning parameters such as pressure, strain, or chemical substitution toward an interaction-driven phase boundary may provide a promising route to maximizing the response.
Read the full abstract
Machine Learning
MUGEN: Generating Unlearnable Graph Examples for Multiple Learning Tasks
Graph data across diverse domains can expose valuable relational information to unauthorized representation learning, creating a pressing need for protection against such misuse. Unlearnable examples offer a data-level defense by perturbing a training release so that models trained on it fail to generalize to clean data. Existing methods generate unlearnable graph examples for only a specified downstream task. Consequently, a release protected against one task may remain learnable for other plausible uses, including node classification, graph classification, and link prediction, which the data owner cannot anticipate. We introduce MUGEN, to our knowledge the first framework for generating unlearnable graph examples that jointly protect all enabled tasks. From one clean dataset, MUGEN produces a single feature-perturbed release that protects every enabled task through a shared GNN encoder and task-specific heads. We devise a Task-Aligned Separability Objective (TASO), which leverages task prediction and classwise separability to strengthen unlearnability and its transfer across GNN backbones and enabled tasks. We further introduce Type-Adaptive Perturbation (TAP), which tailors perturbation optimization to node-attribute type, with direct search over feasible hard flips that accept only loss-improving updates for discrete node attributes and customized gradient-based updates for continuous node features, thereby enabling strong unlearnability across both settings. Experiments across five benchmarks, four backends and three learning paradigms demonstrate that MUGEN generates transferable unlearnable graph examples across GNN backbones and all three tasks, and remains effective under adversarial training and data augmentation.
Read the full abstract
Mesoscale and Nanoscale Physics
Narrow-Sense Type-III Dirac Cones and Additional Flat Lines on a Honeycomb Lattice with Anisotropic Next-Nearest-Neighbor Hoppings
Critically tilted Dirac cones have attracted considerable attention because of their unconventional electronic properties in two-dimensional massless Dirac-fermion systems. Among them, the narrow-sense type-III Dirac cone is characterized by a flat dispersion along the direction connecting the two Dirac points. Using a tight-binding model on a single-orbital honeycomb lattice, we demonstrate that narrow-sense type-III Dirac cones can be realized by tuning anisotropic next-nearest-neighbor hoppings. Type-I and type-II Dirac cones emerge on either side of the critical point, enabling a systematic investigation of the electronic properties across the type-I, narrow-sense type-III, and type-II regimes. We further find that the present model exhibits additional flat lines in momentum space whose energy coincides with the Dirac-point energy at the narrow-sense type-III critical point. These flat lines lead to a pronounced enhancement of the density of states above the Fermi energy and, consequently, the electronic specific heat is substantially enhanced compared with that expected for an ideal narrow-sense type-III Dirac cone. Our results show that the honeycomb lattice with anisotropic next-nearest-neighbor hoppings provides a simple platform for exploring unconventional thermodynamic properties arising from the energy coincidence of narrow-sense type-III Dirac cones and additional flat lines at the Fermi energy.
Read the full abstract
Quantum Physics
The Fermionic Cohomology Problem on the Full Fock Space Is $\mathrm{QMA}_1$-Complete
Fermionic cohomology detects zero-energy states of supersymmetric Hamiltonians. Previous work showed that deciding nonzero cohomology in an input-specified particle-number sector is $\mathrm{QMA}_1$-hard and belongs to $\mathrm{QMA}$. We study total cohomology on the unrestricted full Fock space, where a NO instance must exclude zero-energy states in every sector and in their superpositions. We prove that this global problem is $\mathrm{QMA}_1$-complete. The input differential is an operator that raises fermion number by one and squares to zero on the full Fock space. It is given as an exact list of local fermionic monomials, each involving at most $41$ modes. The reduction encodes each data site by one fermion in a block of modes. Every sector violating this occupation rule has energy at least one. An exact quantum verifier accepts a suitable witness with certainty on every YES instance, establishing containment with perfect completeness. We also prove $\mathrm{QMA}_1$-completeness for the problem in an input-specified particle-number sector. Its hard instances use $30$-mode terms and admit a one-dimensional block-chain realization.
Read the full abstract
Astrophysics of Galaxies
The Stellar Populations of Two Quiescent Low Surface Brightness Dwarf Galaxies in Low Density Environments
Dwarf galaxies located in low density environments that have quiescent stellar populations are extremely rare and the mechanism for quenching their stars is highly uncertain. Here we present two low surface brightness dwarf galaxies with large (>1.2 kpc) effective radii that were previously identified in the Dragonfly Ultrawide Survey for which we have obtained spectra using the KCWI instrument on the Keck II telescope. We derive their stellar populations and star formation histories in several radial bins out to 1.25 effective radii. Globally both galaxies are old (~9 Gyr) and metal-poor ([M/H] ~ -1) and reveal quenching on a relatively rapid timescale (~1 Gyr). We find flat to rising age and metallicity gradients in both galaxies. This is similar to those seen in ultra diffuse galaxies but is in contrast with typical classical dwarf galaxies and predictions from simulations. One galaxy, DFUWS 68, hosts several globular clusters which we find to share similar ages, metallicities and quenching timescales to that of the host galaxy stars.We briefly discuss possible quenching mechanisms and suggest that neither cosmic web stripping nor internal feedback processes alone can explain our results.
Read the full abstract
High Energy Physics (Phenomenology)
Flavor-Violating Higgs and Top Decays in the Type 1B Flavorful Two-Higgs-Doublet Model with a Twist
We study flavor-violating decays of the SM-like Higgs boson and rare top decays in the Type 1B realization of the flavorful two-Higgs-doublet model with a twist, focusing on $h\to sb$, $h\to db$, $h\to uc$, $h\to\mu\tau$, $h\to e\tau$, $h\to e\mu$, $t\to ch$, and $t\to uh$. We present the relevant amplitude and branching-ratio formulae, include one-loop vertex corrections, and impose constraints from Higgs signal-strength measurements, electroweak precision tests, meson mixing, $B\to X_s\gamma$, direct HFV searches and CLFV constraints, including $\mu-e$ conversion. The surviving parameter space favors $\mathrm{BR}(h\to sb)\gg\mathrm{BR}(h\to db)$, $\mathrm{BR}(t\to ch)\gg\mathrm{BR}(t\to uh)$, and $\mathrm{BR}(h\to\mu\tau)\gg\mathrm{BR}(h\to e\tau)\sim\mathrm{BR}(h\to e\mu)$ in the lepton sector. One-loop effects usually preserve these hierarchies, but can modify individual channels through different mechanisms: sizable destructive tree–loop interference can reduce $h\to sb$ enough to yield rare $\mathrm{BR}_{\mathrm{1L}}(h\to db)>\mathrm{BR}_{\mathrm{1L}}(h\to sb)$ points, whereas the large relative corrections found in rare top decays mainly reflect strong SM-like alignment suppressing the tree-level amplitudes.
Read the full abstract
Machine Learning
Verdict Instability of OOD Scores under Reference Resampling
Post-hoc out-of-distribution detectors are fitted on a finite reference set, so every score they produce is an estimate. If we had chosen a different set, some verdicts would have moved. We measure that movement by resampling the reference set and recording the bootstrap standard deviation of the score, which we call verdict instability. It admits a closed form with no fitted parameters. The instability of a verdict is the within-class dispersion of the assigned class along the query's direction, divided by the square root of that class's reference count. That count is what separates verdict instability from the geometry of the score distribution, and it is identifiable only under class imbalance. Instability grows with the local dispersion. Far-OOD queries lie along the low-variance directions of an anisotropic embedding, so the distance-based scores in common use assign their highest values to the verdicts that are most reproducible. Only estimators of local dispersion carry the sign a practitioner expects. We give a rule that predicts this sign for any score from a single label-free correlation, and abstention driven by a wrong-signed score turns out worse than abstention at random on every dataset we test.
Read the full abstract
Probability
Boundary behavior and optimal pointwise Hölder exponent of the total local time of $(1+\beta)$-stable super-Brownian motion
Let $L^x$ be the total local time of one-dimensional super-Brownian motion with $(1+\beta)$-stable branching mechanism, $0<\beta<1$. We prove that $\{x\in\mathbb R:L^x>0\}$ is almost surely a bounded open interval $(\mathsf L,\mathsf R)$ and that \[ h_L(\mathsf L)=h_L(\mathsf R)=1+\frac{2}{\beta} \] almost surely, where $h_L$ denotes the pointwise Hölder exponent. Thus the pointwise $\gamma$-Hölder condition holds at both endpoints for every $\gamma<1+2/\beta$ and fails for every $\gamma>1+2/\beta$. The same conclusions hold under the canonical excursion measure.
Read the full abstract
Computation and Language
SCoNE: Selective Context-aware Neuron Editing for Robust Retrieval-Augmented Generation
Retrieval-Augmented Generation (RAG) is highly sensitive to retrieval noise: when retrieved documents mix informative and irrelevant context, LLMs are easily distracted, leading to hallucinations. To overcome this, we propose SCoNE (Selective Context-aware Neuron Editing), a training-free model editing approach that improves retrieval noise robustness by selectively strengthening context-aware FFN neurons that are identified by both high attribution and high cross-input variability. SCoNE requires only a small number of mining samples, no fine-tuning, and no inference-time overhead. Across various knowledge-intensive question-answering benchmarks and two LLM backbones, SCoNE consistently outperforms competitive baseline methods. Our code is available at https://github.com/HYU-ARK-Lab/SCoNE.
Read the full abstract
Systems and Control
Order-Adaptive Distributed Integral Control
We address a structural tradeoff in distributed dynamic coordination: when the target complexity is unknown, a low controller order saves states but may leave a persistent tracking error, whereas a high order improves tracking but may burden every agent with unnecessary dynamics. To remove this choice without resorting to computationally more involved nonlinear feedback or chattering-prone nonsmooth feedback, we develop an order-adaptive distributed integral controller (OADIC). Specifically, we start with proportional feedback and add integral states only when locally measurable relative errors show that the current order is inadequate. Meanwhile, we organize the candidate controllers in a nested form, thereby preserving the existing states and gains and maintaining continuous control inputs during order transitions. To provide a theoretical basis for this design, we first characterize the consistency of prescribed relative displacements on the augmented agent–target graph. Next, we construct gains that stabilize all admissible fixed-order subsystems and establish a uniform input-to-state stability bound for the variable-dimension closed loop. Furthermore, we prove that OADIC rejects every insufficient order after finitely many decision intervals and explicitly bound the rejection time of the critical order.
Read the full abstract
Number Theory
A unification of Euler-Maclaurin and Euler-Boole summation formulas
In this paper, we establish a summation formula associated with the generalized Apostol-Bernoulli functions. This formula unifies the Euler-Maclaurin summation formula, the Euler-Boole summation formula, and the character analogues of these two formulas. We also apply it to obtain the general power sum formula and the special values of Berndt's generalized $L$-function at non-positive integers.
Read the full abstract
Machine Learning
A Study of Hidden-State Optimization Order in Predictive Coding Networks
Local learning methods offer an alternative to end-to-end backpropagation, but their unstructured local objectives can produce weak feature learning in deep networks. We study whether the order of hidden-state optimization can address this limitation. We propose a boundary-first inference schedule that partitions a model into chunks, first coordinates hidden states at chunk boundaries, and then refines representations within each chunk. We instantiate this schedule in predictive coding networks (PCNs), a local-learning framework in which hidden activities and prediction errors are explicitly exposed during inference. On CIFAR-10, the resulting boundary-first predictive-coding instantiation improves accuracy over standard predictive coding by $9.77\%$ under a standard parametrization and by $5.51\%$ under a $\mu$-parametrization. Diagnostic analyses further show more non-trivial early-layer updates, lower initial-to-final CKA, and more diverse layerwise gradients, consistent with stronger feature learning. These results support boundary-first, chunk-based inference as a practical design principle for predictive-coding training and motivate its study in broader local-learning systems.
Read the full abstract
Computation and Language
Visual Framing for News Stance Detection via Image Generation
Article-level news stance detection aims to identify the perspective of news articles toward social issues. Despite advances in stance detection and its importance for trustworthy media environments, news articles pose distinct challenges because their stances are often implicit, subtly conveyed through journalistic framing, and embedded in long, structurally complex texts. To address these challenges, we introduce VFStance, which leverages visual framing to make implicit stance cues more explicit via image generation. In evaluation experiments, we demonstrate the effectiveness of VFStance over existing methods and the contribution of visual framing to its performance. Finally, a controlled user study (N=200) in a snippet-based news consumption setting further demonstrates that VFStance can make stance signals visually salient and highlights its potential use beyond automated stance detection.
Read the full abstract
High Energy Astrophysical Phenomena
X-ray Flares in Gamma-Ray Bursts at High and Very-High-Energies
A significant fraction of GRB early afterglows exhibit fast and bright X-ray flares, discovered by Swift X-ray Telescope. Their rapid temporal variability and spectral evolution have suggested an internal shock origin, analogous to prompt MeV emission. However, their physical origin and radiation mechanism remain debated. High-energy (> 100 MeV) and very-high-energy (> 30 GeV) gamma-ray observations provide a powerful probe of flare dissipation region, constraining its size, magnetic field strength, and particle acceleration under the synchrotron self-Compton scenario in optically thin relativistic jets. We present a systematic multi-wavelength study of 66 X-ray flares from 47 GRBs observed by X-ray Telescope over 17 years, all within field of view of Fermi Large Area Telescope. We investigate their GeV counterparts and find that only five flares exhibit significant high-energy emission (> 3 sigma). Broadband spectral modeling indicates that this GeV emission is consistent with standard forward shock afterglow. We further investigate correlations between flare spectral properties and energy fluxes at 1 keV, 10 keV, and 1 GeV. Using a synchrotron self-Compton model, we constrain the physical conditions of the emitting region, including magnetic field strength, bulk Lorentz factor, and emission radius. For the most stringent GeV upper limits, we find a magnetic-to-electron luminosity ratio greater than or equal to 1, implying a highly magnetized emitting region. We predict very-high-energy gamma-ray emission from X-ray flares at early (~500s) and late (~5000s) times and assess their detectability with Cherenkov Telescopes. We find that later X-ray flares provide the most promising targets for follow-up observations owing to improved observational accessibility, sensitivity, and reduced response-time constraints of facilities such as Cherenkov Telescope Array Observatory.
Read the full abstract
Computation and Language
Creative Generation via Multi-Agent Debate: Does Debate Suppress Diversity?
Creative generation tasks, such as narrative writing and scientific ideation, demand both high-quality outputs and distinct responses across independent runs to maximize exploration. Multi-Agent Debate (MAD) has shown strong quality gains on factual and reasoning tasks, making it a natural candidate for creative generation. However, we find its convergence-driven design actively suppresses output diversity across independent runs, creating an inherent trade-off with creative tasks. We theoretically show that preserving diversity among agents within each debate session is a necessary condition for achieving diverse outputs across independent runs. Building on this finding, we propose Creative-MAD, which introduces two synergistic interventions to sustain agent divergence. Specifically, Cognitive Lens Assignment counters identity drift by anchoring each agent to a distinct and persistent cognitive mode, while Embedding-based Peer Selection counters majority pull by limiting each agent's context to its most semantically distant peers. Experiments across four creative benchmarks demonstrate that Creative-MAD significantly enhances both lexical and semantic diversity while maintaining MAD's output quality.
Read the full abstract
Robotics
Context-Aware Intelligent Vehicles
Intelligent vehicles increasingly support adaptive applications beyond driving themselves, ranging from context-aware ADAS and automated driving to in-cabin monitoring and fleet management, all under tight requirements on accuracy, latency, cost, and reliability. Meeting these requirements is challenging because vehicles operate in complex, uncertain, and rapidly changing environments while running on resource-constrained computing platforms. This paper argues that context-situational factors that give meaning to sensor signals and constrain decisions-should be treated as a first-class principle for next-generation vehicle systems, and operationalized as a unified, shared state for learning, risk assessment, and closed-loop control across the software stack. We systematically review state-of-the- art (SOTA) context-aware methods spanning (i) environment understanding, (ii) planning and control, (iii) safety and security, and (iv) connected vehicles. Based on a trend analysis of context-aware design, we identify four key technical challenges in building a general contextual engine for future intelligent vehicles: multi-modal context fusion, temporal context modeling, handling rare events, and collaborative context sharing. We hope this survey will motivate the development of robust and efficient context-aware vehicle applications.