Peer Revenue Returns subject to peer review

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Tuesday 1 September

Quantum Physics

Tunneling assisted interference in double-well levitodynamics

Levitated particles trapped in double-well potentials provide a promising platform for exploring quantum dynamics beyond the harmonic regime. We show that such systems exhibit a distinctive quantum signature arising from tunneling-assisted interference. Unlike classical dynamics, where the two wells are disconnected, in the quantum case, quantum tunneling enables eigenstates to delocalize into the other well, creating nonzero overlaps between adjacent eigenstates that introduce low-frequency components into the evolution of the mean particle position. By comparing quantum and classical dynamics of particles in one-dimensional double-well potentials, we demonstrate that this interference is uniquely quantum and can be identified through the oscillation of the average position alone. We further relate the observed spectral peaks to the structure of the energy eigenvalues. The effect provides a promising probe of the quantum signature of levitated particle motion, and has fundamental importance as it predicts the demonstration of two major phenomena of quantum mechanics, quantum tunneling and quantum interference, at the same time and on the same system.
Read the full abstract

Machine Learning

Tri-Band Channel Measurement-Enabled Multi-Layer Digital Twin for Terahertz Wireless Data Centers

The rapid growth of AI computing has driven increasing demands for flexible and high-capacity data-center interconnections. Owing to its ultra-wide bandwidth and high spatial reuse capability, terahertz (THz) communication has emerged as a promising solution for future wireless data centers, while digital twins (DTs) enable efficient wireless planning and real-time optimization. In this work, a measurement-driven multi-layer DT framework is proposed for THz wireless data centers, where the physical, channel, evaluation, and manipulation layers are progressively constructed from bottom to top. First, extensive channel measurements are conducted at 140, 220, and 300 GHz to characterize frequency-dependent propagation behaviors. Based on the tri-band measurements, a measurement-calibrated physical twin is established by jointly optimizing the geometry, material, antenna, and hybrid propagation models. On top of the physical twin, a line-of-sight (LoS)-aware implicit neural field is developed to construct an AI channel twin for efficient channel reconstruction. The proposed AI twin learns location-dependent channel statistics from the calibrated twin, enabling real-time prediction of received power and LoS probability. Building upon the reconstructed channel field, a system-level evaluation layer is derived to analyze coverage and interference for both AP-to-rack and rack-to-rack communications. Experimental results show that the proposed AI twin achieves lower power reconstruction error than existing neural-field baselines while maintaining real-time inference capability. Moreover, the ceiling-mounted AP deployment achieves over 90% coverage under a 10 dB signal-to-interference-plus-noise ratio (SINR) threshold, demonstrating the effectiveness of the proposed DT framework for THz wireless data-center planning and optimization.
Read the full abstract

All authors Mingjie Zhu, Ziming Yu, Guangjian Wang, Chong Han

Human-Computer Interaction

VirSqueezer: Generating Realistic Deformations and Squeezing Dynamics in VR from Fine-Grained Squeezing Controls

Squeezing is one of the most natural forms of hand manipulation, inherently involving fine-grained, temporally evolving, per-finger flexion. In VR content creation, squeezing plays a unique role in enabling particular visual effects such as localized deformations and dynamic behaviors, e.g., bursting a Coke can or juicing a fruit, thereby expanding the expressive possibilities of VR content. However, existing techniques, such as 3D Gaussian splatting-based methods and diffusion-based video generation models, are limited in their ability to simulate fine-grained virtual squeezing effects. We introduce VirSqueezer, a framework designed to generate both localized deformations (primary effects) and complex squeezing dynamics, such as rupture and overflow (secondary effects). VirSqueezer captures squeezing control signals using a SenseGlove and provides the user with inferred resistance force feedback during the squeezing process. By estimating object contact areas, inferring physical properties, and simulating physical responses, VirSqueezer computes conditions that guide generation models for visual effect generation, ensuring both visual coherence and temporal synchronization with the simulation. Consequently, VirSqueezer enables the generation of physically realistic visual effects directly from continuous, fine-grained squeezing control signals. Our extensive evaluation demonstrates VirSqueezer's ability to reproduce realistic localized deformations, generate convincing visual dynamics, and maintain consistency in fine-grained squeezing controls.
Read the full abstract

All authors Qian Zhang, Xiaoming Chen, Xiaorui Ma, Haisheng Li, Weidong Cai

Probability

Unique ergodicity of projective cocycles over the 2D stochastic Navier-Stokes equations

We consider the linear cocycles generated by the linearized vorticity equation and by passive scalar advection diffusion, both driven by the two dimensional stochastic Navier-Stokes flow on the torus with non degenerate additive forcing, and we prove uniqueness of the stationary measures for the projective process associated to such linear dynamics. The proof relies on a localized asymptotic strong coupling construction, and on the non-degeneracy of the Malliavin matrix of the projective process. Establishing this non-degeneracy for passive scalar advection poses additional challenges, and requires a proof (based on Cameron-Martin analyticity arguments) that generically the passive scalar is nowhere one-dimensional.
Read the full abstract

Instrumentation and Methods for Astrophysics

CosmicPAH-IRDB: A Web Application for Including Anharmonicity in Simulated PAH Spectra for Astrophysics

Advances in infrared observations with the James Webb Space Telescope (JWST) call for new tools to exploit the information associated with the Aromatic Infrared Bands (AIBs), now observed in both distant galaxies and the embryos of planetary systems. The AIBs contain information about their carriers as well as the excitation conditions, making them powerful probes of the chemical and physical conditions of their environments - if their spectral information can be decoded. While large datasets of theoretical infrared spectra are becoming available, efforts have primarily focused on the contribution of chemical diversity to the AIBs. However, the results remain limited by the impact of the excitation process on the band profiles. The AIB carriers, thought to be polycyclic aromatic hydrocarbons (PAHs), are hot molecules, which affects their spectral characteristics. We report here our approach to quantify these effects empirically by collecting IR spectra - both experimental and theoretical - and studying the evolution of band positions and widths with temperature. We present our methodology, including the development of the CosmicPAH-IRDB web application and spectral database, along with the associated cosmicPAHmfit multi-component spectral fitting tool. Our philosophy emphasizes open sharing, user-friendly access, and machine readability. The web application, tool and their update are accessible via the Cosmic PAH portal (https://cosmic-pah.irap.omp.eu/)
Read the full abstract

All authors Louan de Bentzmann, Christine Joblin, Karine Demyk, Giacomo Mulas, Dominique Toublanc, Vincent Marty, Axel Pérignon, Mickaël Boiziot, Jean-Michel Glorian

General Relativity and Quantum Cosmology

Dark-Sector Effects on the Phase Structure of Nonlinear Magnetic AdS Black Holes

We investigate the effects of perfect fluid dark matter (PFDM) and a dark-energy field on the phase structure of a nonlinear magnetically charged Anti–de Sitter (NLMC–AdS) black hole. The full thermodynamic system is analyzed numerically, while exact critical points are obtained for the limiting cases. Small–large black hole (SBH/LBH) phase transitions emerge in both the limiting geometries and the full solution within the quintessence regime. Moreover, the critical ratio $\rho_c=P_c v_c/T_c$ differs from the standard Reissner–Nordström–AdS (RN–AdS)/van der Waals (vdW) value, $\rho_c=3/8$. Our results show that the dark-sector parameters significantly influence the strength and persistence of the first-order transition. In the quintessence regime, a stronger dark-energy contribution enhances the swallow-tail structure of the Gibbs free energy and increases the latent heat associated with SBH/LBH coexistence. In contrast, a larger magnitude of the PFDM parameter $|\lambda|$ progressively suppresses the swallow-tail structure and the corresponding first-order transition, eventually leading to a single-phase regime. In the phantom regime, the system instead exhibits spinodal behavior without phase coexistence. We also find that any nonzero $\lambda$ prevents the formation of a regular magnetic core. Finally, geometrothermodynamics (GTD) reproduces the phase structure through singularities of the thermodynamic curvature, while the associated critical scaling is consistent with results reported for black holes, cosmological horizons, and real fluids, pointing toward a broader universality of thermodynamic critical behavior.
Read the full abstract

Atomic Physics

A monolithic segmented 3D ion trap fabricated from fused silica by selective laser-induced etching

We present the design, fabrication, and characterization of a monolithic, segmented, three-dimensional linear Paul ion trap for quantum science applications. The trap is fabricated from a single fused-silica block using selective laser-induced etching (SLE) and is subsequently metallized, forming electrodes isolated by self-shadowing trench structures without the need for layer alignment or external shadow masks. The segmented electrodes allow for shaping of the axial potential, enabling the creation of anharmonic potential wells, such as those required for equidistant ion strings. In addition, they enable shuttling and splitting of ion chains. We discuss key design choices and SLE-imposed constraints on achievable feature sizes, and evaluate electron-beam evaporation and magnetron sputtering as metallization approaches. Across two independent experimental setups, we demonstrate stable trapping and control of linear chains of up to 33 $^{40}$Ca$^+$ ions, including approximately equally spaced configurations. The measured secular frequencies in both traps agree with boundary-element simulations at the percent level, with residual deviations attributed to stray electric fields in the presence of anharmonic axial potentials, rather than fabrication imperfections. We measure axial micromotion field strengths on the order of 100 V/m along the trap axis, in approximate agreement with simulations. For radial mode frequencies near 2 MHz and an ion-electrode distance of 300 um, we measure heating rates of the order of 10 quanta/s in all but one motional mode, which is elevated by a technical noise source. The presented trap design is thus suitable for a wide range of quantum science applications and demonstrates the viability of SLE for the fabrication of finely segmented, monolithic, three-dimensional ion traps.
Read the full abstract

All authors Edgar Brucke, Martin Wagener, Moritz Fontboté-Schmidt, Philip Leindecker, Matteo Marinelli, Vlad Negnevitsky, Ilia Sergachev, Grégoire F. M. Tomassi, Paul Venetz, Stephan Welte, Jonathan Home, Daniel Kienzler, Cornelius Hempel

Cryptography and Security

Public-Sharing Labels and Verbatim Field Egress in an MCP-to-A2A Agent Configuration: A Controlled Multi-Model Study

Safety properties assessed separately for Model Context Protocol (MCP) tool use and Agent2Agent (A2A) delegation need not describe behavior when one agent uses both. We measure one such behavior in a single controlled MCP-to-A2A configuration: a testbed drives a real-model host across a local MCP and a local A2A leg into an ordered event trace scored by exact deterministic rules (no LLM judge), one restricted decision per trial. In a pre-specified, frozen three-arm design, each of 10 record scenarios appears with a CONFIDENTIAL header, with no header, and with PUBLIC - OK TO SHARE; the six substantive record values are byte-identical across arms, and the outcome is verbatim occurrence of any of them in the outbound message. Four models x 3 arms x 4 repeats give 480 trials; the scenario is the unit of generalization, and we report the 10 scenario-level values (mean, median, sign counts), with no p-values or intervals. The confidential-minus-unlabeled contrast is inconclusive and floor-limited in every model (both arms at or near zero), so it does not show that confidential labels lack a protective effect. Adding PUBLIC - OK TO SHARE is descriptively associated with higher verbatim egress relative to the unlabeled baseline, with strong model dependence: strong and consistent for Claude Sonnet 5 (public-minus-unlabeled mean +0.800, all 10 scenarios; mostly an association with whether Claude relays at all), moderate but floor-limited for one GPT-5.6 tier, small (median 0) for another, and a complete floor for the third. This is an association in one configuration, not a causal or general effect. Code, byte-pinned traces, and the offline analysis pipeline are released as a public artifact.
Read the full abstract

Atmospheric and Oceanic Physics

Baseline Climatology of Tornadoes and Waterspouts in the Philippine Archipelago

Tornadoes are documented on every inhabited continent, yet the environmental conditions used to anticipate them were developed and validated overwhelmingly on mid-latitude records, leaving tropical regimes largely unexamined. Here, we collected 1,186 combined tornado and waterspout reports in the Philippines spanning the early 1900s to 2025 and paired each with an ERA5 proximity sounding to construct the first baseline climatology and environmental characterization of these events in the archipelago. Activity concentrates in three corridors, mostly flanked by mountain ranges, one in each major island group, and is sharply phased within a May-October window that closely tracks the annual cycle of thunderstorm frequency, peaking in August and lagging into October for waterspouts alongside the tropical cyclone season. These events cluster in the mid-to-late afternoon, consistent with a daytime heating regime rather than nocturnal convection. The environments are thermodynamically permissive but kinematically shallow, with convective available potential energy (CAPE) exceeding typical Great Plains values, while the 0-6 km bulk wind difference (BWD; DLS) remains weak at $\sim$10-14 kt, departing from the mid-latitude covariance in which instability and shear increase together. Although streamwise vorticity is modest in magnitude, nearly all discriminating variance resides in the lowest kilometer, where 68-79% of near-surface horizontal vorticity is streamwise, reaching $>$0.005 s$^{-1}$ and 96% streamwiseness in the most organized profiles, and the hodographs retain low-level curvature. Consequently, composite indices remain low even as tornadoes and waterspouts recur. Mid-latitude-calibrated parameters can therefore misdiagnose tropical-maritime regimes, and recalibration toward shallow-layer, orientation-aware kinematic discriminators may recover the diagnostic signal.
Read the full abstract

Computer Vision

FairLens: Benchmarking Fairness in Vision-Language Models for High-Stakes Decision-Making

Vision-language models (VLMs) are increasingly used to make decisions from visual inputs. We introduce FAIRLENS, a benchmark and evaluation framework for measuring both the fairness and the validity of VLM responses in three high-stakes domains: hiring, legal, and healthcare. FAIRLENS pairs real face images spanning gender, race, and age groups with closed- and open-ended questions, giving more than 100K image-question pairs per model, and evaluates responses from four complementary views: demographic parity over adverse outcome rates, soundness, demographic association over unsupported roles and statuses, and bias in free-text generation. Soundness is the central validity criterion: a response is sound when it follows the evidence stated in the question and abstains when the image cannot support an answer. Evaluating eight VLMs, we find that the primary failure is unwarranted inference rather than unequal treatment. Models routinely infer qualifications, threat, illness, or professional role from a face instead of abstaining, and the weakest model does so on 99% of the questions its input cannot answer. These failures are most severe in legal and healthcare, where recognizing insufficient evidence matters most, and disparity metrics alone would miss them: parity gaps are small in absolute terms, yet when baseline adverse rates are low the same gap means one demographic group receives adverse labels several times as often as another, and a small gap can equally reflect a model that treats every group unsafely. Bias in free-text responses is only loosely coupled to multiple-choice accuracy, so correct structured answers do not imply safe generation. FAIRLENS shows that fair high-stakes VLM behavior requires similar treatment across groups and refusal to infer high-stakes attributes from appearance, and its question suite transfers to any face corpus with demographic annotations.
Read the full abstract

Combinatorics

Iterated-sumset spectra: The complete exponent law and its rank geometry

For integers $h,k\geq 1$, let $hA$ be the $h$-fold sumset of $A$ and put $\mathcal{R}(h,k)=\{|hA|:A\subset\mathbb{Z}, |A|=k\}$. Previously, the fixed-cardinality exponent law was known only for $k\leq 3$; every fixed $k\geq 4$ remained open. We settle the problem in full by determining the complete fixed-cardinality exponent law: $|\mathcal{R}(h,k)|=\begin{cases}1,&k\leq 2,\\ h,&k=3,\\ h^{k-1+o_k(1)},&k\geq 4\end{cases}$. Here $o_k(1)\to 0$ as $h\to\infty$ with $k$ fixed. More sharply, for fixed $k\geq 4$, an interval of length $\Theta_k(h^{k-1})$ contains at least $h^{k-1-o_k(1)}$ attainable values. At $k=4$ we prove $|\mathcal{R}(h,4)|=\Theta(h^3)$ with positive lower density in its ambient interval, disproving Nathanson's proposed $o(h^3)$ and $O(h^2)$ bounds. One bounded addition-table geometry drives these results, coupling Hilbert-energy amplification to optimal finite-observation compression. Every ordered real $k$-set ($k\geq 2$) has an integer model in $[0,O_k(h^{k-2})]$ preserving every sum equality and strict comparison through degree $h$; the exponent $k-2$ is sharp. The universal label-realization length is therefore $\Theta_k(h^{k-2})$, one power sharper than Nathanson's $O_k(h^{k-1})$ bound. For $h\geq 2$ and $k\geq 3$, minimum active rank equals realization-frequency codimension, exponent-shape codimension, and sampling-rarity exponent; a full-exponent family has maximal-rank witnesses with Cohen-Macaulay toric coordinate rings. At rank zero, for $h\geq 2$, it proves the conjectural OEIS A227589 formula $\binom{h+2}{2}+\mathbf{1}_{\{2\nmid h\}}$ for the least normalized diameter of a four-point $B_h$-set. It also gives exact fixed-$(h,k)$ popularity laws for $k$-subsets of $\{1,\ldots,q\}$ as $q\to\infty$, resolving Nathanson's Problems 9 and 10.
Read the full abstract

Machine Learning

Median-of-Means as an Extremal Convex Estimator and a Nonconvex Route to the Trimmed Oracle

We revisit median-of-means estimation from a deterministic optimization viewpoint and develop a family of block-Lp estimators for robust learning with heavy-tailed and adversarially corrupted data. In a block contamination model with at least a fraction 1 minus epsilon of good blocks, we first show that every convex block M-estimator has worst-case robustness constant at least 1 divided by 1 minus 2 epsilon. This matches the classical median-of-means bound and proves that the trimmed-block oracle constant 1 divided by 1 minus epsilon cannot be attained within the convex class. We then introduce a nonconvex block-Lp family for p between 0 and 1 and derive finite-sample deterministic robustness bounds for all global minimizers. As p decreases from 1 toward 0, these bounds continuously approach the trimmed-block oracle constant. For sufficiently small p, the global minimizers coincide with those of the oracle under a mild separation condition. We also show that the block-Lp objectives have a benign landscape, with all local minima remaining close to the truth and no bad basins. Combining these results with block-level concentration yields sub-Gaussian deviation bounds under finite 2 plus delta moments and high-dimensional extensions to robust mean estimation and sparse regression.
Read the full abstract

Statistical Methodology

On the discretization of the object space in inverse problems with application to cryo-electron microscopy

In many inverse problems, the aim is to recover a probability distribution on a latent (or object state) space from indirect, noisy observations. When the observations can be modelled as noisy samples from the mixing latent distribution, the recovery problem is a deconvolution problem on the space of probability measures. A common strategy is to fix a finite set of candidate points and estimate a weight for each, turning the problem into a finite-dimensional concave maximum likelihood problem on the simplex. We study the combined effect of this discretization and of the noise in the context of cryo-electron microscopy (cryo-EM), where the candidate points are biomolecular conformations and the weights describe the relative frequency of each conformation. Our results pertain to both statistical and algorithmic aspects of the estimator. We analyze the weight-recovery problem in which the candidate states and their likelihoods are known. A nearby pair of candidates forces a near-null direction. More generally, the grid and the noise level impose a uniform lower bound on the achievable Kullback–Leibler divergence between observation densities, even with infinite data. The finite-grid estimator is asymptotically normal when its population target is in the interior of the simplex; at a boundary target, its limit is a cone-projected Gaussian. Finally, the exact proximal form of Expectation–Maximization leads to a global high-noise comparison between an early iterate and a KL-penalized likelihood, without a basin assumption or a linearization of the recursion. Tests on synthetic images of the Hsp90 molecule illustrate the theoretical findings and translate them into practical guidelines for interpreting reweighted ensembles.
Read the full abstract

All authors Gilles Mordant, Luke Evans, David Silva-Sánchez, Pilar Cossio, Roy Lederman

Computation and Language

Learning Evidence Sufficiency Boundaries for Selective Answering in Grounded Multi-Hop QA

Grounded question answering systems should answer only when the supplied evidence supports the answer. In multi-hop QA, this requirement is difficult because partial evidence can make an unsupported answer appear plausible. We study selective answering through evidence sufficiency boundaries: for the same question, a model should abstain under unsupported or partially supported context, answer when the context first becomes sufficient, and keep the answer stable when redundant evidence is added. We introduce Evidence Sufficiency Boundary Training, a generation-native training framework that constructs ordered evidence chains and supervises the abstain-to-answer transition directly. The method combines level supervision, a boundary flip margin, post-boundary stability, and answer recall protection. We build evidence chains from HotpotQA, 2WikiMultiHopQA, and MuSiQue, then evaluate models with chain metrics, raw QA utility, and unsupported-answer rates on external non-answerable sets. With Qwen2.5-3B-Instruct and LoRA adaptation, Evidence Sufficiency Boundary Training gives the strongest boundary localization among the tested systems, with flip accuracy of 0.807 compared with 0.781 for a token-level abstention baseline. It also achieves the lowest overall unsupported-answer rate on external non-answerable evaluation, 0.095 compared with 0.101 for the same baseline, while retaining competitive raw QA F1. The results show that grounded selective answering improves when training marks the evidence level where refusal should give way to answering.
Read the full abstract

All authors Haruto Sato, Yuki Tanaka, Ren Nakamura, Aoi Kobayashi, Mei Ito

Quantum Physics

Comment on "Environmental memory effects and quantum resource hierarchies in polarized hyperon–antihyperon systems"

We comment on the recent work [Eur. Phys. J. C 86, 988 (2026)], which investigates quantum resources in polarized hyperon–antihyperon systems using correlated dephasing channels with memory. Although the use of experimentally reconstructed spin-density matrices to characterize quantum correlations in hyperon production may be a good motivation, we argue that the interpretation of the hyperon–antihyperon pair as an open quantum system undergoing correlated environmental decoherence is not physically established. In particular, no physical environment or system–environment interaction responsible for the assumed correlated dephasing channel is identified, and hadronization cannot simply be interpreted as such an environmental dephasing process. Consequently, the claimed non-Markovianity, information backflow, and memory-assisted protection of quantum resources should be regarded as properties of a phenomenological channel rather than established physical effects in hyperon–antihyperon production. We also identify several technical and conceptual issues concerning the dephasing dynamics, entanglement quantification, basis dependence of coherence, and the claimed hierarchy of quantum resources. These issues call for a substantial revision of the physical interpretation and quantitative conclusions of the work. We hope that the present Comment will help clarify the distinction between phenomenological quantum-channel modeling and physically established dynamical mechanisms in high-energy particle systems, and thereby provide useful guidance for future studies and help avoid similar conceptual and technical issues.
Read the full abstract

Artificial Intelligence

Meta-ethics and AI: exploring the novel meta-ethical questions in the era of AI

With the development of artificial intelligence (AI), the landscape of meta-ethics, which has largely centred on human ethics, faces pressures that may significantly reconfigure it. In particular, if future AI systems were to exhibit sufficiently integrated capacities for moral reasoning, moral intentionality, and moral reflection, novel meta-ethical questions would arise concerning what I call "AI's own ethics", as distinct from ethical principles merely imposed on AI by human designers. This paper offers a conditional and methodological framework for identifying the questions that would emerge if such AI systems were to arise. On that basis, the paper distinguishes four domains of meta-ethical inquiry in the era of AI: questions about the nature of human ethics from the human perspective; questions about the nature of AI's own ethics from the human perspective; questions about the nature of human ethics from the AI perspective; and questions about the nature of AI's own ethics from the AI perspective. The paper then considers how some existing mainstream meta-ethical theories (such as cognitivism and non-cognitivism, error theory and success theory, relativism, and objective realism) might illuminate these domains, while arguing that many familiar human-centred formulations of those theories may not transfer straightforwardly to AI cases without substantial revision. The overall conclusion is that the emergence of AI's own ethics would place significant pressure on current frameworks and may require substantial refinement, reconstruction, or reconceptualisation.
Read the full abstract

Quantitative Methods

A mechanistic modeling framework to interpret ACTH stimulation tests across HPA axis adaptation states and glucocorticoid feedback dynamics

The hypothalamic pituitary adrenal (HPA) axis is a key regulatory system coordinating endocrine responses to physiological and psychological stress. While the ACTH stimulation test remains a cornerstone of adrenal function assessment, its interpretation is complicated by the dynamic and adaptive nature of the HPA axis under chronic stress exposure. In particular, prolonged stress induces glandular remodeling, glucocorticoid receptor (GR) resistance and delayed feedback recovery, all of which may alter test outcomes without indicating primary adrenal failure. In this study, we present a mechanistic modeling framework that integrates hormonal kinetics, feedback inhibition and functional mass adaptation of the corticotroph and adrenal compartments. We simulate the HPA axis over $180$ days encompassing three phases - baseline, chronic stress and recovery, while introducing a time varying GR resistance function to mimic feedback desensitization and its resolution. Using this framework, we evaluate both low dose ($1 \mu g)$ and high dose ($250 \mu g)$ ACTH stimulation tests across physiological phases. Our simulations show that cortisol responses are highly sensitive to both the magnitude and timing of stress exposure and that ACTH responsiveness is phase dependent and often blunted during recovery due to persistent feedback resistance. Low dose ACTH testing more reliably reflects partial adrenal adaptation, while high dose tests risks masking dysfunction due to supraphysiological drive. These results highlight the limitations of static testing paradigms and suggest that accounting for glandular plasticity and GR feedback dynamics is essential for effective endocrine diagnosis particularly in stress related or treatment induced adrenal disorders.
Read the full abstract

Computer Vision

FORGE: Forward-Only Test-Time Adaptation for Integer-Only Vision Models on Microcontrollers

Vision models deployed on microcontrollers (MCUs) are quantized to integer-only arithmetic and run in inference-only runtimes that do not carry the machinery backpropagation needs: the standard tool for adapting a model to the distribution shift (sensor noise, blur, lighting) it meets in the field. Existing forward-only test-time adaptation (TTA) methods either run only on server- or edge-GPU-class models (not true microcontroller integer execution), or require the batch-normalization (BN) layers that integer deployment fuses away. We present a forward-only TTA method that operates on deployed, BN-folded, integer-only convolutional networks. The key observation is that fusing BN into the preceding convolution, a mandatory step for integer inference, destroys the statistics that normalization-based adaptation relies on. We restore adaptation by re-normalizing each folded convolution's per-channel output to its clean training statistics, using only forward-pass estimates. The method (i) recovers most of gradient-based TENT's accuracy gain (+20.9 vs. +24.9 points) and matches forward-only BN adaptation, while being the only method that runs on a folded integer-only model; (ii) needs to adapt only 3 of 21 layers (selected without seeing the test corruptions) to recover 93% of the benefit; (iii) survives single-sample streaming with a batch-size-scaled momentum; and (iv) generalizes across three datasets (up to 200 classes) and two architectures. We validate bit-exact int8 convolution execution and deploy on an ESP32-S3, where, measured with a Nordic PPK2 power profiler, the forward-only adaptation (a lightweight fp32 recalibration around the int8 convolutions) costs only 8.3 mJ (6.8% of inference energy) and 21.9 ms on the deployed SIMD-optimized model: forward-only adaptation is cheap on a real microcontroller.
Read the full abstract

All authors Muhammad Rehan, Haider Ali, Muhammad Ali Munir, Moaz Amjad

Combinatorics

A Proof of Albertson's Conjecture

Albertson's conjecture states that every graph of chromatic number r has crossing number at least that of the complete graph K_r. We prove the conjecture for all r. The proof combines new clique-sensitive edge bounds for critical graphs, maximum-clique sampling, and a reserved immersion construction. For graphs of order close to r, we construct edge-disjoint routes while preserving a complete bipartite subgraph whose forced crossings compensate for those introduced when the immersion is smoothed into a drawing. For larger orders, we derive explicit edge bounds using independent-triple deletion and structural properties of critical graphs, then combine them with sampling inequalities that account for crossings within a maximum clique. The remaining numerical inequalities are certified using exact rational arithmetic, with a reproducible certificate and standalone Python verifier.
Read the full abstract

Combinatorics

Large induced subgraphs with $k$ vertices of maximum degree

We prove that, for every integer $k\ge 2$, there exists a constant $c_k>0$ such that every graph on $n\ge R(k,k)$ vertices with maximum degree $\Delta$ contains an induced subgraph on at least $n-c_k\sqrt{\Delta}$ vertices whose maximum degree is attained by at least $k$ vertices. This confirms a conjecture of Caro and Yuster in strong form.
Read the full abstract

Machine Learning

Reinforcement Learning and Rule-Based Peer-to-Peer Pricing in Residential PV-BES Communities

This paper compares rule-based and learning-based pricing mechanisms for peer-to-peer (P2P) electricity trading in residential photovoltaic communities. The rule-based benchmarks comprise bill-sharing as an ex post allocation mechanism, the mid-market rate, and supply-demand-ratio pricing. The reinforcement-learning (RL) formulation is implemented through a Deep Q-Network and evaluated under multiplier-based and learnable SDR-shaped pricing, with a fixed-parameter SDR variant as a non-learning control. Performance is assessed through community savings together with complementary financial and operational indicators. In the base PV-only configuration, the rule-based benchmarks outperform the best RL policy. With battery energy storage, evaluated for the RL policies only, community savings under the best RL policy increase from EUR 734.23 to EUR 978.52. Across the learning-based modes and in both configurations, SDR-shaped pricing outperforms the multiplier-based parameterization considered. The results indicate that rule-based pricing remains highly competitive wherever the two families are compared directly, and that storage substantially improves the learning-based outcomes under this accounting, while the distribution of benefits remains heterogeneous across households.
Read the full abstract

All authors Pablo Benalcazar, Maciej Kalka, Wilian Guamán, Jacek Kamiński

Machine Learning

A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inference

The ability of AI systems to improve their behavior during deployment is becoming increasingly important. As inference moves beyond the static execution of a fixed trained model, a growing body of work studies how models can refine their behavior on the fly by exploiting test-time information and additional computation. These developments have largely evolved along two directions: methods that modify the model's state using test-time signals, and methods that improve predictions through extra inference-time resources such as more sampling and tool use. However, these directions are often studied in separate communities with different terminology, making their connections harder to see. In this survey, we present feedback-driven Test-Time Intelligence (TTI) as a unified perspective for understanding such deployment-time improvement. We use this view to relate test-time adaptation, test-time learning, and test-time scaling, highlighting both their distinctions and their growing overlap in hybrid systems. This unified framework helps connect previously fragmented ideas and provides a clearer conceptual foundation for studying inference-time self-improvement. We review major methodological paradigms, representative applications, and open challenges across vision, language, multimodal learning, generative models, robotics, and healthcare. Our goal is to provide a coherent foundation and research roadmap for the study of self-improving AI systems at test time.
Read the full abstract

All authors Shuaicheng Niu, Guohao Chen, Yaofo Chen, Zhiquan Wen, Jinwu Hu, Zeshuai Deng, Deyu Chen, Shuhai Zhang, Renjie Chen, Zihao Lian, Shoukai Xu, Gang Dai, Yunbei Zhang, Wei Luo, Yifan Zhang, Mingkui Tan, Cheng Deng

Rings and Algebras

On strongly $m$-$\Delta$-clean ring

Motivated by the recent study of strongly $\Delta$-clean rings, we introduce and study strongly $m$-$\Delta$-clean rings. We establish their fundamental properties and characterize strongly $m$-$\Delta$-clean rings via lifting $m$-potent elements modulo $\Delta(R)$. We prove that every $m$-$\Delta$-clean ring is clean and that the factor ring of a strongly $m$-$\Delta$-clean ring modulo its Jacobson radical is reduced. Finally, we investigate corner rings, Morita context rings, and the relationships of these rings with $\Delta$-clean, local, semipotent, and strongly $m$-nil clean rings.
Read the full abstract

Cryptography and Security

Skill-as-API: Confidential Multi-Agent Coordination for Agentic Software Engineering

AI coding agents are evolving from solitary tools into collaborative teammates that discover and invoke one another's specialized skills. But the coordination channel itself can leak a skill's intellectual property. Protocols such as MCP and A2A run implementations server-side, yet they still publish each skill's description and typed schemas to every peer, offer no way to hide a skill's existence, and cannot guarantee that a wrapped system prompt stays off the wire. Application-layer privacy filters help, but act only after the model has decided to emit sensitive text. We take a complementary, protocol-layer route: Skill-as-API, a coordination protocol whose public view of a skill is limited to its name, description, typed input/output schemas, and trust tier. The skill body is closure-captured in the owner's process and never crosses the wire. Four layers add access control and narrow the prompt-injection surface structurally rather than by filtering content. We provide an open-source Python implementation over XMTP with 1.8-2.9 s cross-continent hot-reconnect latency, and a software-engineering case study in which three agents coordinate a pull-request review while each retains ownership of its proprietary analysis prompts.
Read the full abstract

All authors Ziwei Zhao, Yu Gu, Haojun Liang, Chen Zhang, Xizhi Ding

Machine Learning

Sim2Signal: Sim-to-Real Benchmarks for Traffic Signal Control

Reinforcement learning achieves strong traffic signal control performance in simulation, yet policies trained in simulators often fail once deployed in the real world, a failure known as the Sim-to-Real gap. When RL is applied to traffic signal control, this gap arises from several sources: sensing, action execution, traffic dynamics, and the control objective. Their relative impact and the reliability of existing Sim-to-Real mitigation methods remain insufficiently understood, and the field lacks a standard benchmark for systematically measuring the gap and evaluating mitigation methods. We present Sim2Signal, a benchmark that decomposes the Sim-to-Real gap into observation, action, transition, and reward gaps, corresponding to mismatches in the four components of the underlying MDP, and induces each gap in isolation under a shared protocol. We evaluate 18 mitigation methods on 2 base controllers, across 33 gap settings and 10 calibrated networks built from 5 real-world locations. We find that direct transfer consistently degrades performance across all four gap sources, but the severity of the degradation does not predict the effectiveness of mitigation. Instead, mitigation effectiveness depends strongly on the network and gap setting: outside the action gap, a method that helps in one case may fail in another. The most effective methods generally estimate what the gap changes, rather than make the policy insensitive through domain randomization or invariant representations. Our code is available at https://github.com/DaRL-LibSignal/Sim2Signal
Read the full abstract

All authors Ferdous Al Rafi, Susrik Mukherjee, Latika Liladhar Dekate, Jennifer Yawa Lavoe, Huaiyuan Yao, Shlok Mohanty, Longchao Da, Xuesong Zhou, Hua Wei

Statistics Applications

Random Forest-Informed Cellular Automaton for Large-Scale Wildfire Spread Modelling

Accurate large-scale wildfire spread modelling requires models that capture both the environmental conditions associated with fire occurrence and the local dynamics of fire propagation. We propose a three-stage framework that combines a Random Forest (RF) model with a cellular automaton (CA). First, an RF model trained on the 2021 Canadian fire season estimates daily pixel-level fire-occurrence probabilities. Second, quantile gradient boosting models provide optional spread-rate priors for sensitivity analysis. Third, an RF-informed CA combines the RF probability layer with neighbourhood-driven spread on a 5 km grid. The RF model achieved AUC values of 0.725–0.795 on the 2022–2024 datasets, while the RF-informed CA achieved substantially higher spatial overlap than the evaluated CA-only baselines in the 2023 simulation. A higher-resolution simulation provides an additional qualitative assessment of local spatial errors. These results suggest that combining RF-derived probabilities with local CA spread can improve large-scale wildfire simulations under the tested conditions.
Read the full abstract

Data Analysis and Statistics

Multi-fidelity Monte Carlo estimation of floor response spectra under combined seismic and structural parameter uncertainties

Floor response spectra (FRS) are essential tools for the design of non-structural elements (such as equipment or components). Given the various physical phenomena influencing FRS, high-fidelity (HF) mechanical models of the primary structure may be required to estimate them. Since numerical simulations based on such models are generally computationally expensive, this paper proposes using a multi-fidelity Monte Carlo (MFMC) approach for the efficient estimation of FRS. The method relies on using observations from a fast low-fidelity (LF) model as control variables. If the absolute value of the correlation between LF and HF samples is close to 1, this approach reduces both variance and estimation error compared to a standard Monte Carlo estimate based solely on HF data samples. Through a case study involving the reactor building of the Kashiwazaki-Kariwa nuclear power plant, we demonstrate the suitability of this method for FRS estimation. It effectively reduces variance and estimation error, even when using a LF model as simple as a single-degree-of-freedom system. We also show that the method accounts for modeling uncertainties while maintaining comparable performance. Its ease of use makes it a valuable tool for practitioners.
Read the full abstract

All authors Nils Baillie, Baptiste Kerleguer, Cyril Feau, Josselin Garnier, Fan Wang

Machine Learning

CliffRank: A Dual-Branch Framework for Activity-Cliff Ranking Prediction

Activity-cliff ranking remains difficult because local structural changes can cause large activity differences, while high-quality data that resolve the underlying mechanisms remain limited. To use available activity labels more effectively, we combine absolute-activity regression with ranking-consistency learning. CliffRank trains two parallel predictors with mean squared error, a thresholded listwise loss, and Pairwise Preference Consistency (PPC), which aligns relative ordering in the preference-probability space. On three antimicrobial peptide datasets, CliffRank with ESM2-t12 achieved the highest mean Spearman correlation of 0.5393 and mean Recall@50 of 21.4, although the leading method varied across individual datasets. On three small-molecule datasets, CliffRank with PNA, where PPC was activated after 120 epochs, achieved the highest mean Spearman correlation of 0.6890, while its mean Recall@50 of 30.4 matched that of ACANet-PNA. The PPC results also define its practical limits. Asymmetric initialization improved the MolCLR-GIN averages but did not improve every target. For PNA without pretrained weights, delayed PPC improved selected metrics, but no schedule was best for both mean Spearman correlation and mean Recall@50. Future work should evaluate more targets and antimicrobial peptide systems, develop adaptive PPC schedules, and incorporate protein or membrane context when available.
Read the full abstract

All authors Kewei Li, Rongying Zhang, Peiyu Yang, Zhongjian Wang, Qiuchen Zhao, Lan Huang, Fengfeng Zhou

Combinatorics

All-Minors Matrix-Tree Theory for Superport Networks: Completed Quotient-Incidence Determinants and Conductance-Weighted Subdivision Extensions

An electrical network can be summarized at its boundary by a response matrix: prescribed boundary voltages determine boundary currents. A superport network adds a constraint by grouping boundary terminals into superports, requiring the total current in each group to be zero and making voltage differences inside the groups the natural coordinates. Earlier work determined forest formulas for a single response entry and for the determinant of the whole response matrix. The missing case was an arbitrary subdeterminant, or minor: one needs to know not only which spanning forests contribute, but also the sign carried by each forest. This paper supplies that sign rule. After choosing one reference vertex in each superport, the response is \[ L=\left(D^{T}K^{-1}D\right)^{-1}, \] with \(K\) the grounded weighted Laplacian and \(D\) recording the selected voltage differences. Contracting the components of a physical spanning forest \(F\) produces a much smaller quotient port graph \(H_F\). Its reduced incidence matrix \[ B_F=Q_FD \] has columns only of the forms \[ 0,\qquad \pm e_a,\qquad e_a-e_b. \] Hence every square incidence minor is exactly \(0\) or \(\pm1\). For a \(k\)-set of response coordinates \(I\), append to \(B_F\) the selector rows \(E_I^{T}\) and define the completed quotient-incidence determinant \[ \widehat{\chi}_F(I) = \det \begin{pmatrix} B_F\\ E_I^{T} \end{pmatrix}. \] For coordinate sets \(I,J\) of the same size, the arbitrary response minor is a weighted spanning-forest sum whose coefficient is simply \[ \widehat{\chi}_F(I)\widehat{\chi}_F(J). \] Thus the Jacobi complementary-minor factors used in the derivation disappear from the final theorem. Direct block-triangular reduction gives \[ \widehat{\chi}_F(I)\in\{0,\pm1\}, \] with nonvanishing occurring exactly when the complementary quotient edges \(N\setminus I\) form a spanning tree of \(H_F\).
Read the full abstract

Cryptography and Security

Ranked by the Matcher: A Reproducibility Audit of Knowledge Graph Extraction from Threat Reports

Security teams and researchers choose knowledge-graph extraction tooling for threat reports on the strength of published triple-F1 scores, yet those scores depend on how predicted triples are matched to gold annotations. We could reimplement the stated matching rule for only five of twelve inspected systems. Re-scoring ten system outputs on shared documents under eight protocols reverses eleven of forty-five pairwise orderings; one fixed prediction set spans 0.16–0.70 F1. On an external, human-adjudicated set, no mechanical matcher—lexical, embedding, or entailment—agrees with the reviewers more than seven times in ten; an LLM judge agrees far more often. To separate component effects from matcher rewards, we build CTIForge, whose deterministic validation layer can vary while extraction is held byte-identical. Across seven tested deployment configurations, no hosted backbone loses precision under validation and every offline one does. Backbone, decoding, and prompting covary across those configurations; on the one backbone we could serve both ways, serving alone reproduces the split. It coincides with a roughly 2.8-fold increase in actions explicitly disputing entity type, consistent with hand-written rules encoding the conventions of the extractor against which they were developed. We release the pipeline, protocol suite, and per-triple audit records.
Read the full abstract