Machine Learning
BLOKS: Scaling Kernel Solvers with Sketched Block Preconditioning
Giordano Vitale, Giacomo Meanti, Marco Rando et al. (+2 more)
Kernel Ridge Regression (KRR) is a theoretically sound
algorithm for nonparametric learning, which is also often used to investigate properties of deep learning. However, the exact KRR formulation incurs time and
memory costs preventing applicability to large
datasets, motivating the development of approximate solvers. These
algorithms rely on (i) iterative methods, to avoid the computational burden of direct solutions; (ii) sketching or batching, to reduce the problem size; and (iii) preconditioning, to accelerate convergence. While crucial for fast convergence, preconditioning incurs
memory costs that can be managed by lowering the number of free parameters of the model, potentially at the expense of expressivity and hence accuracy. To address this challenge, we propose BLOKS, a Conjugate Gradient-based solver for Nyström-sketched KRR built around a novel preconditioner that combines a block Jacobi approximation with sketching. The BLOKS preconditioner has
memory cost independent of both the
dataset size and the number of parameters, allowing scaling to large models. We provide a
memory-efficient, GPU-accelerated implementation and show that BLOKS achieves excellent accuracy while retaining favorable time/
memory complexity. Indeed, on standard large-scale
datasets used to
benchmark kernel solvers, BLOKS achieves comparable or better performances than state-of-the-art methods.
Read the full abstract ▾
All authors Giordano Vitale, Giacomo Meanti, Marco Rando, Blanca Ayuso de Dios, Lorenzo Rosasco
Optics
Bottom-Up Surface-Enhanced Raman Scattering for the Detection of Micro-Objects and Globular Analytes
Andrey K. Sarychev, Milena S. Shestopalova, Denis S. Korzhov et al. (+7 more)
Surface-enhanced Raman scattering (SERS) is typically achieved by depositing the analyte on a metal or
semiconductor substrate. We propose a bottom-up approach, enabling SERS by depositing a metal on the surface of the object being studied. A Mie-theory analysis of electric-field enhancement in dielectric core-shell particles is presented. A correlative approach combining SERS, atomic force microscopy, and scanning electron microscopy is employed to model the conditions required for sensitive SERS detection and spectral analysis of polystyrene microspheres (PMSs) as models of globular and vesicle-like objects. Based on these findings, we propose a SERS-active system that provides strong signal enhancement (up to four orders of magnitude for a single globular object) while maintaining spectral fidelity, where PMSs act both as model analytes and as integral components of a metal-dielectric resonator.
Read the full abstract ▾
All authors Andrey K. Sarychev, Milena S. Shestopalova, Denis S. Korzhov, Andrey V. Ivanov, Igor V. Bykov, Konstantin N. Afanasev, Nikita N. Osipov, Dmitry V. Basmanov, Vladimir A. Oleinikov, Konstantin E. Mochalov
Geophysics
Is Earthquake Replenishment Sufficient for Unbiased b-Value Estimation?
Herbian Hokky Prabowo and Giuseppe Petrillo
Short-term aftershock incompleteness biases Gutenberg-Richter b-value estimates because many small earthquakes remain undetected immediately after large events. Two strategies address this problem. Earthquake replenishment reconstructs missing events to restore a more complete catalog, whereas robust positive-difference
estimators infer the b-value directly from incomplete observations. Whether these approaches provide equivalent information on the magnitude distribution remains open. We address this issue using synthetic aftershock sequences generated with the Epidemic-Type Aftershock Sequence (ETAS) model and degraded through a time- and magnitude-dependent incompleteness model. Incomplete catalogs are replenished using AutoBETA, and b is estimated using the standard maximum-likelihood, b-more-positive, and b-more-incomplete
estimators. Replenishment reconstructs the overall structure of the missing-event population and substantially improves conventional maximum-likelihood estimates, although small residual biases remain. By contrast, both positive-difference
estimators recover the imposed b-value directly from incomplete synthetic catalogs. Applications to the Ridgecrest, Landers, and Hector Mine sequences show that b-more-incomplete is less sensitive to replenishment than b-more-positive. Ten independent replenishment seeds indicate that seedto-seed variability is small relative to differences among
estimators. Replenishment and robust b-value estimation therefore address different aspects of catalog incompleteness: replenishment provides a more complete aftershock representation, whereas robust
estimators are preferable when accurate b-value estimation is the primary objective.
Read the full abstract ▾
Quantum Physics
Wideband microwave resonator for NV center vector magnetometry
S.V. Bolshedvorskii, E.A. Koreshin, V.V. Soshenko et al. (+10 more)
The nitrogen-vacancy (NV) color center in diamond has found applications in magnetometry across various geometrical scales. In particular, vector magnetometry using hand-held devices with low power consumption is of interest for mobile and geological applications. However, such a device requires efficient driving of the four different orientations of the NV center, thus placing demands on a broadband microwave resonator that provides a uniform and highly insensitive microwave
magnetic field at the resonance frequencies of the NV centers. Here, we propose a unique planar resonator design that enables efficient driving of all four orientations of NV color centers over a 100 MHz range with only 0.5 W of microwave input power and 250 mW of
laser power, allowing vector
magnetic field measurements. Using the planar resonator to drive NV centers in diamond, a compact vector magnetometer was fabricated and experimentally studied. The sensitivity of the device is about \[{1.2\text{ nT}}/{\sqrt{\text{Hz}}}\;\]. While this is not a record sensitivity for diamond-based magnetometers, this device occupies its own niche as a compact hand-held magnetometer.
Read the full abstract ▾
All authors S.V. Bolshedvorskii, E.A. Koreshin, V.V. Soshenko, I.S. Cojocaru, S.M. Drofa, P.G. Vilyuzhanina, N.I. Salangin, A.M. Kozodaev, A. Chernyavskiy, A.N. Smolyaninov, P.V. Kapitanova, S.Ya. Kilin, A.V. Akimov
Functional Analysis
On the simultaneous zero inclusion property and characterization of Inner product space
Souvik Ghosh
We show that if a
Banach space $X$ is smooth and almost locally uniformly rotund and if satisfies the following \[0\in W(T) \implies 0\in W(T^{-1})\] for all invertible bounded linear map on $X$ then Birkhoff-James orthogonality in $X$ is symmetric. This result, therefore, characterizes the inner product spaces out of all smooth and almost locally uniformly rotund spaces having three or more dimension via simultaneous zero inclusion property, introduced by Bračič et al. [J. Math. Anal. Appl., 449 (2017) 1413-1423]. Further, this gives answers to some open questions asked by E. Poon [J. Math. Anal. Appl., 534(2024), 128052].
Read the full abstract ▾
High Energy Physics (Theory)
On the quantum statistical relation and the covariant phase space Hamiltonian
José Luis V. Cerdeira and Tomás Ortín
We show how the covariant phase space
Hamiltonian can be constructed in theories in which some of the fields have gauge freedoms. The addition to the action of boundary terms that guarantee the well-posedness of a variational principle and its exact gauge invariance for the prescribed boundary conditions turn out to be essential in this construction. The same ingredients are essential in the use of the action in the framework of Gibbons and Hawking's Euclidean approach to
Quantum Gravity and we show how the on-shell action constructed with a Lagrangian with those properties can be expressed in terms of the covariant phase space
Hamiltonian. We compute the
Hamiltonian of several theories, one of which has Chern–Simons terms, using our prescription and show that, in black-hole backgrounds, the
Hamiltonian gives directly the
quantum statistical relation including terms related to non-gravitational charges and chemical potentials, as expected.
Read the full abstract ▾
Robotics
EndoRSI: Physical-Evidence-Driven Recursive Improvement of Endoscopic Robot Policies
Chi Kit Ng, Yidong Zhang, Tianchun Wu et al. (+4 more)
Artificial intelligence is beginning to automate scientific discovery, yet physical
robot-policy research still relies on scientists and engineers to interpret experimental failures and decide how controllers should be revised. We present EndoRSI (Endoscopic Recursive Self-Improvement), a human-governed, self-improving robotic research system in which the embedded EndoForge coding
agent iteratively develops and refines endoscopic
robot policies using evidence from physical experiments. Physical execution remains isolated within a safety-constrained runtime under human authorisation. In motorised endoscopy, validation evidence first caused the
agent to reject an initial action representation, while subsequent physical experiments falsified an offline-supported insertion strategy by revealing hazardous closed-loop behaviour absent from the source data. Failed physical rollouts were then incorporated into policy revision; insertion success subsequently increased from 2/21 in the first generation to 17/21 in the fourth. The frozen final policy retained competence across visual and hardware shifts and was the best performing autonomous policy across six independent ex-vivo porcine specimens. These results show that physical
robot interaction can serve as experimental evidence for
agent-driven
robot policy research.
Read the full abstract ▾
All authors Chi Kit Ng, Yidong Zhang, Tianchun Wu, Siu Hing Lui, Hanlin Zhao, Huxin Gao, Hongliang Ren
Computation and Language
From Structure to Validity: Mechanism-Guided Generation and Refinement of LLM Rubrics
Feifei Qian, Xuan Chen, Lixin Cui et al. (+6 more)
Rubrics have been widely used to evaluate and optimize
language models on open-ended tasks. However, existing work primarily focuses on generating better rubrics, with limited understanding of what makes rubric design effective. In this work, we conduct a systematic mechanistic analysis of rubric design across criterion organization, criterion operationalization, and score aggregation. Our analysis shows that structural organization enables more comprehensive criterion construction, semantic specification alone does not ensure effective criterion-level judgments, and sophisticated score aggregation provides limited additional gains. Guided by these findings, we introduce Adaptive Rubric Tree Search with Hybrid refinement (ARTS-Hybrid), which combines adaptive tree construction with semantic-first, response-calibrated refinement. Experiments across
benchmarks show that ARTS-Hybrid improves both end-to-end evaluation performance and intrinsic rubric quality. Further experiments suggest that, beyond evaluation, rubrics generated by ARTS-Hybrid provide effective signals for model optimization.
Read the full abstract ▾
All authors Feifei Qian, Xuan Chen, Lixin Cui, Chaoyang Mei, Fanlin Meng, Shuyao Xiao, Ming Cui, Junxi Yin, Lu Bai
Combinatorics
On the fault-tolerant metric dimension of maximal outerplanar graphs
Carmen Hernando, Montserrat Maureso, Mercè Mora et al. (+1 more)
Given a graph $G$, a set $W$ of vertices of $G$ is a resolving set if, for any pair of vertices $z_1$ and $z_2$ of $G$, there is a vertex $w\in W$ such that $d(w,z_1)\ne d(w,z_2)$, where $d(u,v)$ denotes the length of a shortest path between the vertices $u$ and $v$. The metric dimension of a graph $G$, $\beta(G)$, is the minimum cardinality of a resolving set of $G$. A resolving set $W$ is fault-tolerant if $W\setminus \{w\}$ is a resolving set for any $w\in W$. The fault-tolerant metric dimension of a graph $G$, denoted by $\beta'(G)$, is the minimum cardinality of a fault-tolerant resolving set of $G$. In this paper, we study the fault-tolerant metric dimension problem in maximal outerplanar graphs. In particular, we show that there are exactly two maximal outerplanar graphs with $\beta'(G)=3$. We also show that $4\le \beta'(G) \le 6$ for any maximal outerplanar graph $G$ of order at least 7 and metric dimension 2. Moreover, we establish the upper bound $\beta'(G) \le \left\lceil \frac{2n}{3} \right\rceil$ for maximal outerplanar graphs $G$ of order $n\ge 5$, by building in linear time a fault-tolerant resolving set of cardinality $\left\lceil \frac{2n}{3} \right\rceil$.
Read the full abstract ▾
All authors Carmen Hernando, Montserrat Maureso, Mercè Mora, Javier Tejel
Robotics
PartiGen: Part-Aware Representations for Language-Conditioned Humanoid Motion Generation
Chuxuan Huang, Zhanxiang Cao, Tianyue Zhan et al. (+14 more)
Translating
language into
humanoid motion requires coordinating body parts, determining motion duration, and producing references suitable for
robot execution. We introduce PartiGen, a part-aware latent
diffusion framework that incorporates anatomical structure into a shared whole-body motion representation. Part-specific
attention processes anatomically masked frame features, and its outputs are fused to combine local limb information with whole-body context. Conditioned on text and motion history, the model generates motion primitives autoregressively. Forward-kinematics supervision encourages kinematic consistency, while a learned end-of-sequence predictor selects a frame-level endpoint after generation. On HumanML3D retargeted to the Unitree G1, PartiGen achieves a segment-level FID of 3.165, a 16.8% reduction over TextOp, and improves R@1 by 1.4 percentage points. In the duration study with 16-frame primitives, endpoint prediction reduces mean absolute duration error from 12.471 to 3.080 frames compared with a fixed-length cutoff. We further evaluate
robot execution using a pretrained, frozen GRIT tracker: 409 of 500 generated references (81.8%) pass the posture-based stability screen in MuJoCo. Following simulation screening and manual selection, 26 of 27 attempted motions are judged successful on the real G1.
Read the full abstract ▾
All authors Chuxuan Huang, Zhanxiang Cao, Tianyue Zhan, Haoyang Li, Yang Zhang, Liyun Yan, Chenlin Ming, Keqi Zhu, Yizhi Chen, Shengcheng Fu, Zihao Zhou, Xinyi Peng, Kaixuan Zhang, Boyang Xing, Jiawei Bao, Yang Yang, Yue Gao
Cryptography and Security
Towards Safe World Models: Risks, Defenses, and Evaluation
Xiaoma Wang, Litian Zhang, Xinze Shi et al. (+11 more)
World models support planning, control, learning from imagined experience, and model-mediated evaluation by predicting action-conditioned consequences. Their safety cannot be inferred from predictive accuracy alone: errors in observations, state and
memory, dynamics, evaluation, execution, or feedback can propagate into unsafe decisions and persist through updates. This work organizes the literature on world-model safety around a six-stage lifecycle covering data and observations, state and
memory, dynamics and imagination, evaluation and planning, action and control, and feedback and update. We synthesize non-
adversarial failures, attacks, defenses, evaluation protocols,
benchmarks, and deployment conditions, and distinguish evidence directness from outcome level. The current evidence corpus contains 163 cited works. We use the framework to identify where claims are supported, where attribution remains unresolved, and which measurements are needed to connect model-level behavior to closed-loop and operational safety.
Read the full abstract ▾
All authors Xiaoma Wang, Litian Zhang, Xinze Shi, Zhengxuan Hu, Zihan Yang, Mingqi Ren, You Wu, Bingyu Yan, Zelong Yu, Zhenjie Su, Yuming Shang, Xi Zhang, Chaozhuo Li, Philip S. Yu
Group Theory
On $p$-groups with one nonlinear codegree
Jiwei Zheng and Heng Lv
For an irreducible character $\chi$ of a finite group $G$, its codegree is defined by $cod(\chi)=|G:\ker\chi|/\chi(1)$. In this paper, we characterize fnite $p$-groups whose nonlinear irreducible characters all have the same codegree.
Computation and Language
Overcoming Data Imbalances with Expert Selection Biases
Ayush K. Tarun, Clara Meister and Antoine Bosselut
Mixture-of-Experts (MoE) models have become the standard architecture for frontier LLMs. The multiple parallel experts in each
transformer block theoretically enable expert specialization according to categories such as domains,
languages, or modalities during pretraining. However, when data is imbalanced across categories, minority categories can become concentrated in a small fraction of the expert pool or spread inefficiently across them, effectively limiting their available capacity. In this work, we demonstrate that category-conditioned expert-selection biases can preferentially route
tokens from different categories toward specialized expert pools. By applying biases based on input category, we provide a simple way to induce soft category-aligned specialization. This intervention encourages categories toward distinct expert subsets while retaining cross-category sharing when useful. More importantly, it enables the over-allocation of expert capacity to categories that would otherwise be disadvantaged by data scarcity. We observe performance improvements across modalities, low-resource
languages, and code in models up to 10.6B parameters. At our smallest and largest model scales, these gains are equivalent to allocating 18.5% and 2.7% additional baseline training
tokens, respectively. Overall, our results show that expert biases can serve as a lightweight control for category-level expert allocation, making room for low-resource categories while retaining standard MoE inference.
Read the full abstract ▾
Machine Learning
Opening the Black Box of Classifier-Free Guidance via Information Bottleneck
Jiayang Gao, Tianyi Zheng, Jiayang Zou et al. (+5 more)
Classifier-free guidance (CFG) is a standard technique for improving conditional generation in
diffusion models, yet its activation schedule is usually selected heuristically. In practice, most existing CFG-based methods rely on a pre-defined activation policy throughout inference, while recent dynamic variants often depend on hand-designed schedules without a principled theoretical explanation of when conditional information is most useful. In order to understand how the informational relevance between conditioning inputs and the score function evolves over the course of generation, we explore when guidance should be activated from an information-theoretic perspective. In this work, we formulate the
diffusion trajectory as a time-dependent information bottleneck and show that the effectiveness of conditional information is not uniform throughout sampling, but becomes significantly stronger after a critical stage characterized by a Fisher-information balance condition. Motivated by this insight, we propose a training-free dynamic CFG strategy for inference. Rather than prescribing a fixed schedule in advance, our method constructs a
stochastic process from conditional and unconditional score statistics and activates guidance when the estimated guidance gain crosses a prescribed threshold. This yields an adaptive activation rule that aligns condition injection with the underlying generation dynamics, while also recovering interval-based guidance as a limiting special case. Extensive experiments across diverse architectures and tasks, demonstrate that our method is effective, robust, and broadly applicable. Beyond empirical improvements, our framework provides an interpretation of dynamic guidance methods and a principled foundation for inference-time control in conditional
diffusion models.
Read the full abstract ▾
All authors Jiayang Gao, Tianyi Zheng, Jiayang Zou, Fengxiang Yang, Luyao Fan, Lv Tang, Bo Li, Jia Wang
Hardware Architecture
Anechoic: Onsager-Corrected Parallel Annealing for Sub-0.1 ms Dense Max-Cut on an FPGA
Seungki Hong, Kyeongwon Jeong and Taekwang Jang
Ising machines solve combinatorial optimization problems by annealing a network of
spins. On dense problems, the least expensive hardware step updates all
spins at once:
stochastic cellular automata (SCA) decide every
spin in parallel and read couplings only for the
spins that flip. Such synchronous updates create an echo: the previous state of each
spin returns to it through its neighbors. On K2000, a Max-Cut
benchmark on a 2,000-node complete graph, two-thirds of all flips undo the same
spin's flip one step earlier. This paper presents Anechoic, an FPGA Ising machine that subtracts this echo from every decision. Dynamical mean-field theory identifies the echo as an Onsager reaction; its one-step strength is the number of
spins whose flip probability is strictly between 0 and 1, divided by twice the temperature. The decision logic already flags these
spins, so the correction costs one population count per step; a temperature-scaled form needs none. An RTL engine executes a dense 2,000-
spin step in 18.1 cycles plus 0.14 per flip, so every saved step and flip saves time. Both forms find better K2000 cuts at every step budget than five published binary-
spin algorithms at their papers' settings. Twelve engines at 250 MHz on an AMD Alveo V80 reach a cut of at least 33,000 with 99% probability in 0.05 ms, 5.2 times faster than the fastest published hardware, at 5.4 mJ per solution; with 2-bit couplings, the temperature-scaled form is also faster than uncorrected SCA on 49 of 51 G-set graphs. On an NVIDIA GH200 GPU, the same engine is 2.3 times slower and needs 8.0 times more energy per solution, though its 240 concurrent anneals give 2.3 times the throughput. As a GlobalFoundries 22 nm ASIC at 1.25 GHz, twelve engines would reach the target in about 10 microseconds, 26 times faster than the fastest published hardware. The code is available at https://github.com/skethz/anechoic.
Read the full abstract ▾
Cryptography and Security
Secrets in Preamble: Practical and Robust PHY Information Hiding for Wi-Fi
Rui Pan, Guanxiong Shen, Linning Peng et al. (+2 more)
Physical layer (PHY) information hiding refers to the concealment of secrets within PHY waveforms without compromising the integrity of the primary communication link, thereby enabling critical applications such as covert communication, PHY steganography, and watermarking. The widespread deployment of Wi-Fi has motivated substantial research interest in hiding secrets within Wi-Fi waveforms. However, existing Wi-Fi information hiding approaches suffer from two limitations: restricted steganographic capacity and insufficient robustness against environmental variations, both of which severely constrain practical deployment. This work proposes to embed secrets within the Wi-Fi legacy short training field (L-STF). The designed method preserves the autocorrelation properties of Wi-Fi L-STF, ensuring that packet detection and coarse frequency offset compensation functionalities remain uncompromised. The secret
embedding and extraction processes employ a pair of neural networks designed to learn optimal
embedding rules and maximize steganographic capacity. We developed a prototype system and conducted comprehensive experiments across both indoor and outdoor environments. Experimental results show that our method can reliably embed 46 secret bits within Wi-Fi packets across diverse channel conditions, achieving significantly enhanced steganographic capacity and environmental robustness. The hardware prototype study validates the feasibility of the proposed Wi-Fi information hiding scheme.
Read the full abstract ▾
All authors Rui Pan, Guanxiong Shen, Linning Peng, Xinyu Qi, Junqing Zhang
Probability
Discrete total variation under the multinomial, Part~1: expectation under the uniform null
N. Banić and N. Elezović
We study the expectation of the discrete total variation statistic $d=\sum_{i=1}^{n-1}|X_{i+1}-X_i|$ of counts $X_1,\dots,X_n$ over $n$ ordered bins with sample size $N$, under the uniform multinomial null. In the notation used throughout the series, a reference distribution $\mathbf r$ defines $d_{\mathbf r}(x)=\|x-N\mathbf r\|_V$; here $\mathbf r=\mathbf u$ is uniform and we write simply $d$. We obtain an exact single-sum formula, a closed form in central trinomial coefficients valid for every $n\ge2$, with one-term evaluations at $n=2,4$ and a complete D-finite characterization for $n=3$. The per-edge expectation $W_N=E|X_2-X_1|$ satisfies an exact three-term recurrence, from which we derive its complete $1/N$ expansion with the $n$-independent constant $E(d^\circ/\sqrt{Nn})\to 2/\sqrt\pi$ for the cyclic statistic $d^\circ=d^\circ_{\mathbf u}$, independent of the number of bins; the complementary sparse regime (fixed $N$, $n\to\infty$) is treated in a separate paper. Conditioning on a two-bin total reduces each edge to a symmetric random walk; as the subsequent parts of the series will show, this method carries the analysis to non-uniform reference laws.
Read the full abstract ▾
Number Theory
Short Recurrences and Straight-Line Programs for Primes
Joseph M. Shunia and Mihai Prunescu
We give a short autonomous recurrence that generates every odd
prime, in increasing order, from the single initial value $8$. Its integer state determines both its output and its next update: the lowest set bit gives the next
prime, while the highest set bit determines the sieve period. The same state also determines the following
prime gap. A related recurrence represents the
prime-counting function as the $2$-adic valuation of a single integer. Independently, a direct packing of a finite divisor sieve yields fixed arithmetic constructions for the $n$-th
prime, the
prime-counting function, and the twin-
prime counting function. We express these constructions as straight-line programs using addition, truncated subtraction, multiplication, integer division, exponentiation, and remainder. Counting operations with reuse of computed values gives programs of lengths $166$, $132$, and $167$, respectively; one autonomous recurrence update uses $99$ operations. Explicit arithmetic subroutines remove all auxiliary operations, including bitwise AND. The programs unfold into fixed arithmetic terms without variable-length sums, products, or searches.
Read the full abstract ▾
Distributed Computing
Stacked Intelligent Metasurface-Based MultiModal Feature Fusion for Collaborative Inference in Edge Networks
Yushi Wang, Zhengxin Yu, Guhan Zheng et al. (+3 more)
In navigating the challenges of computational and communication overhead within the 6G-era Collaborative Inference (CI), this paper explores the feasibility of employing Stacked Intelligent Metasurface (SIM) to perform CI, termed SIM-CI. The SIM functions as a physical computing unit, receiving multi-user Over-The-Air (OTA) aggregated features within the same spectrum. It thus executes the necessary spatial-frequency combining and inference in a single physical wave propagation.SIM-CI also incorporates a tri-heterogeneity robustness adapter to address the network heterogeneity, such as multi-modal data and user inference models, as well as a quality-aware contrastive approach that align features from different modalities and extraction levels into a unified feature space. Considering the heterogeneous and dynamic channels, we further integrate Channel State Information (CSI) as an additional input into the adapter design, enabling users' early perception of potential channel states. Moreover, to maximally mitigate the impact of channel variations, we design a new hybrid beamforming network that combines the robustness of model-driven pre-equalization with the adaptability of data-driven learning. Experimental results demonstrate the effectiveness of our SIM-CI compared to digital-domain inference, achieving 98.6% accuracy to conversional digital-domain CI approaches under various heterogeneous conditions, while reducing computational energy by 25% and requiring 75% fewer transmission time slots.
Read the full abstract ▾
All authors Yushi Wang, Zhengxin Yu, Guhan Zheng, Haris Pervaiz, Zheyi Chen, Shunqing Zhang
Computers and Society
Adversarial Attacks Reveal Divergent Deception Competence in Humans and Generative Language Models
Lucca Pfründer, Riccardo Loconte, Bruno Verschuere et al. (+1 more)
Machine learning models typically outperform humans in deception detection. A threat to machine learning models is their vulnerability to
adversarial attacks: subtle yet targeted perturbations of input data that trick a model into misclassification. Yet, this also provides an opportunity: understanding how humans versus machines construct attacks offers new insights into second-order beliefs about deception. In a preregistered experiment (n = 501), we compare the effectiveness, efficiency, and nature of a paraphrasing attack by humans and generative
language models on an automated deception classifier. We find that humans were more effective attackers when rewriting truthful statements to be deceptive, while the generative
language model was more effective and efficient at making lies sound like truths. Computational text analysis suggests that this is because humans make targeted use of deception cues, while generative
language models, instead, apply generic changes that signal truth. The failure to mimic deception is two-sided: while it exposes a blind spot in the machines increasingly used to detect lies, it also means that
language models might facilitate successful lying.
Read the full abstract ▾
All authors Lucca Pfründer, Riccardo Loconte, Bruno Verschuere, Bennett Kleinberg
Statistics Theory
Finite-Sample Bounds for Fréchet Means of Dependent Data in Metric Spaces with Curvature Bounded Above
Zoraida F. Rico and Christof Schötz
We prove finite-sample risk bounds for empirical quadratic and power Fréchet means in geodesic metric spaces with
curvature bounded above for observations with arbitrary joint law. We compare the sample with exchangeable copies in which one observation is redrawn independently of the original sample and the others move to preserve the joint law. These displacements control the error, with no
entropy or dimension assumptions. For quadratic Fréchet means, this yields a general “coupling bound” in any separable complete $CAT(\kappa)$ space, where for $\kappa>0$ the observations must lie in a closed convex set of circumradius below $\pi/(2\sqrt{\kappa})$. This bound is exact on the real line for nonnegatively correlated Gaussian observations and for comonotone functions of a common factor. We apply it in two settings. Under local dependence, the resulting risk bound grows linearly with the average size of the dependency neighborhoods. Under transport sensitivity and for bounded metric variances, summable sensitivities of the conditional laws to a change in one observation preserve the parametric mean squared error rate, while nonsummable
polynomial decay gives explicit slower rates. We verify transport sensitivity for stationary reversible Markov chains with Wasserstein-contractive kernels and for bi-Lipschitz images of Gaussian sequences. In the more challenging setting of heavy-tailed, dependent data with possibly infinite variance, we consider power Fréchet means of order $\alpha\in(1,2)$. Under nonpositive
curvature, we bound their $L^\alpha$ risk assuming only finite $\alpha$-th moments.
Read the full abstract ▾
Software Engineering
AgentSWE: Can Coding Agents Build the Agent You Actually Want?
Jiahao Wang, Hongjin Qian, Yuyang Hu et al. (+2 more)
As the production of software passes into the hands of software itself, coding
agents are increasingly commissioned to build
agents fitted to their users' own workflows. This delegation carries a quiet assumption, that the delivered
agent will faithfully serve the requirement it was built for, rather than merely appear to. No existing
benchmark tests that assumption. We introduce AgentSWE, a
benchmark for
agent software engineering spanning three stages of the
agent development lifecycle: Creation of a complete
agent from a natural-
language requirement, Editing of production
agent codebases, and Optimization toward a measurable target. Each of its 25 tasks is framed as a commission rather than an exam: the builder iterates against a handful of illustrative cases, while acceptance exercises the frozen submission on held-out cases designed to span the full requirement, grading behavior against requirement-derived rubrics under hard pass/fail gates or, for Optimization, against the source
benchmark's own objective. Evaluating a range of coding
agents across models and harnesses, we find that none comes close to delivering what a requirement asks for: the gap between what a builder demonstrates and what it delivers is the norm, not the corner case. An accompanying empirical analysis of the failure trajectories identifies recurring failure modes, among them overfitting to the visible cases, hollow runtimes behind code that looks complete, and unsupported claims of completion, each a signal by which coding
agents can learn to build the
agents they are asked for.
Read the full abstract ▾
All authors Jiahao Wang, Hongjin Qian, Yuyang Hu, Jiajun Zhang, Zheng Liu
Functional Analysis
An Improved Inequality between the von Neumann-Jordan and James Constants
Zhiyao Fang and Qi Liu
We provide a strengthened answer to the questions posed by Alonso, Martín, and Papini [Studia Math. 188 (2008), 135–150] concerning the comparison between the von Neumann–Jordan and James constants . The classical inequality $\CNJ(X)\leq J(X)$ between the von Neumann–Jordan and James constants was established by Takahashi and Kato [J. Math. Anal. Appl. 359 (2009), 602–609], Wang [Proc. Amer. Math. Soc. 138 (2010), 695–701], and Yang and Li [Appl. Math. Lett. 23 (2010), 277–281], while related sharp estimates were obtained by Kato and Takahashi [Rend. Circ. Mat. Palermo (2) Suppl. 82 (2010), 75–91]. We sharpen this inequality to $\CNJ(X)\leq\JB(X)$, where $\JB(X)$ is the Birkhoff–James orthogonal James constant introduced by Baronti and Papini [Constr. Math. Anal. 5 (2022), 37–45]. Our proof employs optimal distance ellipses and alternating contact directions.
Read the full abstract ▾
Quantum Physics
System-Design Optimization of a Single-Transmitter, Single-Fiber QKD-based ITS Communication Link
Shivang Srivastava, Guillaume Ricard, Yves Jaouën et al. (+2 more)
Deploying QKD-based secure communication at scale requires reducing duplicated optical hardware while preserving the rate of protected user data. Earlier studies established the physical feasibility of carrying CV-QKD and classical signals over shared infrastructure, motivating a quantitative comparison between optical hardware savings and the OTP-encrypted rate. We perform this comparison by defining the secure communication rate (SCR) as the rate jointly supported by secret-key generation and classical transport, and using it to optimize a shared discrete-modulation CV-QKD link. The model combines a binomial 64-QAM security bound with a trusted receiver, protocol traffic, spectral leakage - optical power leaking into the
quantum receive band - and Raman scattering. At 25 km, the asymptotic model gives an optimized SCR of 12.079 Mbps, compared with 12.095 Mbps for a separated-link reference, while reducing the transmitter and fiber counts from two to one each in the modeled direction. Increasing channel separation suppresses spectral leakage, but the resulting SCR gain diminishes once receiver bandwidth limits the rate. Under the adopted receiver-noise law, increasing the bandwidth budget from 350 to 750 MHz raises the projected 25-km SCR to 24.535 Mbps despite higher trusted electronic noise. For the studied metropolitan link, sharing saves optical resources with little rate loss, while receiver-bandwidth improvements provide greater benefit than further spectral separation. Together, these results suggest an integration path: share the transmitter and fiber, limit unwanted classical power in the
quantum receive band, and increase usable receiver bandwidth where the noise budget permits.
Read the full abstract ▾
All authors Shivang Srivastava, Guillaume Ricard, Yves Jaouën, Matteo Schiavon, Romain Alléaume
Functional Analysis
Second quantization and compactness of induced Gaussian quantum Markov semigroups
Zheng Li
We characterize compactness of the
Hilbert-Schmidt
semigroups induced by finite-mode Gaussian
quantum Markov
semigroups with a faithful normal invariant state. A
unitary second-quantization representation for every $ s $-
embedding shows that compactness at one positive time, compact resolvent, drift stability, and uniqueness of the normal invariant state are equivalent. These conditions imply immediate trace-class membership for all
embeddings. We derive exact trace-norm and
Hilbert-Schmidt norm formulas, an
embedding-independent trace identity, and the generator spectrum with algebraic multiplicities. Examples show that compactness does not require the existence of a spectral gap and that a spectral gap need not imply compactness when the invariant state is not unique.
Read the full abstract ▾
Signal Processing
ECGWmark: Deep Watermarking of Electrocardiogram Signals With Error-Correcting Codes
Kuan-Yu Chen, Kuan-Chen Wang, Yu Tsao et al. (+1 more)
Wearable electrocardiogram (ECG) recordings pass through multiple networks and systems before automated interpretation, making patient metadata vulnerable to loss. Digital watermarking embeds this metadata directly into the signal. However, conventional ECG watermarking methods fail under temporal desynchronization, and waveform similarity does not guarantee downstream diagnostic preservation. This article presents ECGWmark, a deep watermarking framework for single-lead ECG. A fully convolutional generator embeds the payload without signal quantization, a detector decodes the message frame by frame and verifies watermark presence, and a frozen arrhythmia classifier penalizes changes that would alter diagnostic predictions. To bridge the gap between bit accuracy and whole-message recovery, payloads are protected by extended Hamming codes, validation-selected interleaving, and soft-decision decoding. On the MIT-BIH Arrhythmia Database, ECGWmark maintains over 90% bit accuracy under time shifts of 10 to 50 samples, where transform baselines fall to chance level. With the coding layer, the protected 16-bit payload achieves 99.86% exact recovery on clean segments and sustains over 90% recovery under time shifts, 10% sample deletion, and 10% packet loss. Inter-patient evaluation confirms improved ventricular-beat sensitivity with preserved macro-F1. Requiring only 0.56 million parameters and about 15~ms per segment on a desktop CPU, ECGWmark highlights the need to assess physiological watermarking through whole-message recovery and diagnostic consistency.
Read the full abstract ▾
All authors Kuan-Yu Chen, Kuan-Chen Wang, Yu Tsao, Jian-Jiun Ding
Artificial Intelligence
SWE-Critix: Post-Training Generalist Verifiers for Assessing Coding-Agent Trajectories
Jiajun Hu and Chun Yong Chong
Coding
agents are increasingly deployed for end-to-end software engineering tasks, where a readily available oracle evaluator that defines the success criteria is often absent. This motivates verifications for coding
agents that can reliably judge and explain their behavior. However, verification is non-trivial as coding
agents are increasingly capable of solving long-horizon software engineering tasks. Their execution trajectories can be extremely long, containing hundreds of iterative interactions with code repositories and external tools. Existing verifiers either rely on general-purpose LLMs without specialized training or train dedicated critics that predominantly focus on producing scalar rewards, highlighting the relatively limited exploration of training-based generalist verification. We introduce SWE-Critix, a post-training pipeline for training generalist verifiers of coding-
agent trajectories. Given a task and its execution trajectory, the learned verifier predicts the completion label and generates a natural-
language analysis that explains its judgment. Our pipeline combines supervised fine-tuning (SFT) with reinforcement learning (RL) and trains a series of 30B models on trajectories from CodeForge-Preview. On a held-out test set, SWE-Critix improves classification accuracy from 57% for the base Qwen3-30B-A3B
reasoning model to 66% after SFT and 78% after RL. We release our models,
datasets, and training code to facilitate further research.
Read the full abstract ▾
Numerical Analysis
On the Convergence of the Mesh-Free Discretization of a Non-Linear Bond-Based Peridynamic Model with a Bond-Breakage Criterion
Lukas Pflug, Michael Stingl and Max Zetzmann
We show the convergence of a commonly used mesh-free discretization of a dynamic non-linear bond-based peridynamic model with irreversible damage implemented via a bond-breakage criterion for a fixed horizon $\delta >0$. The model chosen here is a mathematically well-posed regularization of a discontinuous peridynamic operator often found in the literature, which assumes that fracture results from strain in tension only. We explicitly include the one-point quadrature used in this scheme in the analysis. The convergence is with respect to the natural norm of the solution space given by $C^1([0,T],L^\infty(\Omega)^d)$, and therefore $L^\infty(\Omega)^d$ in the spatial coordinates. This result heavily constrains the spatial regularity of the continuous solution which we prove afterwards. More specifically, for locally $L$-Lipschitz continuous input data, local Lipschitz bounds are derived for the displacement, regardless of damage. This is consistent with interpreting the peridynamic model used here as an approximation to a truly discontinuous fracture model with a length-scale parameter given by the horizon $\delta>0$. We note that we do not deal with the limit of the solutions as $\delta \searrow 0$.
Read the full abstract ▾
Artificial Intelligence
AlphaNIST: Emergent Assembly Skills from Binary Task Rewards
Octi Zhang, Iretiayo Akinola, Rosario Scalise et al. (+12 more)
The NIST Assembly Task Board #1 remains a challenging
benchmark for learning-based methods in robotic manipulation. Over short horizons, it demands precise, contact-rich control of small, tight-fitting parts. Over long horizons, it requires recovery and coordination across extended action sequences. Demonstrations, task decomposition, and dense progress rewards have enabled research advances by providing guidance on how to assemble. But how much of the strategy must be specified? The progression from AlphaGo to AlphaGo Zero showed that RL at scale can discover superhuman strategies without prescribing how to win. AlphaNIST asks whether a sparse binary task reward and task diversity admitted by this task-agnostic reward can similarly let a
robot discover how to assemble. In simulation, we train a single policy from scratch across all 20 assembly tasks on the NIST Assembly Task Board #1, reaching 98.2% overall training success without demonstrations, progress shaping, scripted primitives, or per-task tuning. The framework also trains policies to complete ten assembly pairs in a single episode without a prescribed order, with task success rewarded only when all ten pairs are simultaneously complete. The learned policies discover dexterous and planning-like behaviors: choosing grasps for later assembly, using surrounding geometry to reorient parts, and preparing one part, assembling another, then returning to finish the first without an explicit planner. For
videos, see our website: https://alpha-nist.github.io/
Read the full abstract ▾
All authors Octi Zhang, Iretiayo Akinola, Rosario Scalise, Mateo Guaman Castro, Patrick Yin, Tobias Widmer, Miles Macklin, Chris Amevor, Dylan Turpin, Dirk Van Gelder, Jan Carius, Ankur Handa, Gavriel State, Yunrong Guo, Yashraj Narang
Multimedia
EmoGavel: Grounding-Aware Reinforcement Learning with Verifiable Rewards for Multimodal Emotion Preference Learning
Zilong Huang, Kong Aik Lee, Zezhong Jin et al. (+2 more)
Multimodal emotion preference learning uses pairwise comparisons to identify which open-ended
emotion descriptions align more closely with human judgments. However, existing methods primarily focus on the final preference decision and provide little explicit supervision for localizing and correcting erroneous content. To address this limitation, we propose EmoGavel, a Grounding-Aware reinforcement learning framework with Verifiable rewards for
multimodal Emotion preference Learning. We first introduce grounded supervised fine-tuning (Grounded SFT), which exploits controlled-error construction information to jointly learn preference prediction, error localization, repair, and missing-evidence recovery. We then optimize these verifiable outcomes with grounding-aware group relative policy optimization (GA-GRPO), combining preference-gated grounding rewards with dynamic zero-variance rescue to prioritize preference correctness and recover learning signals from all-wrong groups. Experiments on EmoPrefer and controlled-error evaluation sets show that EmoGavel improves both natural preference prediction and controlled-error discrimination while maintaining strong error-grounding performance. Our code and model checkpoints will be released once the paper is accepted.
Read the full abstract ▾
All authors Zilong Huang, Kong Aik Lee, Zezhong Jin, Man-Wai Mak, Tatsuya Kawahara