Number Theory
Coxeter Symmetry in Hypergeometric Functions and Elliptic Integral Moments
Dianbin Bao
We develop a hypergeometric framework for
elliptic integral moments organized around Mishev's completed Saalschützian $L$-function and its $W(D_5)$
symmetry. The Bailey–Mishev very-well-poised ${}_7F_6(1)$ representation places complementary
elliptic integral moments and harmonic hypergeometric sums in a common parameter space. We first give collision-based hypergeometric derivations of the three complementary-modulus moment families involving $K'^2$, $E'K'$, and $E'^2$. The $E'^2$ identity is also recovered independently from a Barnes-contiguous relation mirroring the differential system of the
elliptic integrals. We then derive a hypergeometric evaluation of the harmonic-series representation of the four-Bessel moment $s_{4,0}$. At symmetric parameter points, Coxeter
symmetry forces selected Taylor coefficients to vanish along
symmetry-adapted paths, producing families of harmonic and special-value identities involving values of the Riemann
zeta function and the same Bessel period. The resulting picture shows that these
elliptic-moment, harmonic-sum, and special-value identities arise from a common $W(D_5)$
symmetry of generalized hypergeometric functions.
Read the full abstract ▾
Artificial Intelligence
A Computer Vision Approach to Visual Fraud Detection in Phishing Websites Using YOLOv8
Basil Sajid Shaikh
Phishing remains one of the most common vectors for financial and identity fraud, and most detection systems still rely on inspecting a page's URL, HTML markup, or domain registration history. These signals are easy for an attacker to rotate or obfuscate, and they say very little about what actually convinces a victim to hand over a password or a card number: the way the page looks. This paper describes a visual, image-based approach to phishing detection that treats a rendered webpage the same way a human eye would, as a picture that either matches a trusted brand or doesn't. A YOLOv8 convolutional neural network was trained to classify full-page website screenshots as phishing or legitimate based on layout, logo placement, color scheme, and login-form structure, rather than on text extracted from the page. The system reached 92% classification accuracy on a held-out test set, processed a single screenshot in roughly 100 milliseconds, and, after a round of data augmentation aimed specifically at lighting, compression, and scaling variation, cut the false-positive rate by 11% relative to the pre-augmentation baseline. The paper walks through the
dataset construction, the augmentation strategy, the model architecture and training setup, and the resulting performance, and closes with a discussion of where this kind of visual detector fits alongside, rather than instead of, existing URL- and content-based defenses.
Read the full abstract ▾
Machine Learning
LTV-CTDNet: Compositional Turning Decomposition for Short-Term Turning-Movement Forecasting
Md Atiqur Rahman Mallick, Kamrul Hasan and Robert T. White
Short-term turning-movement forecasts can support signal control and corridor operations, but unconstrained neural networks may produce physically impossible negative counts or outputs that are not explicitly tied to an approach-demand total. This study introduces the Linear Temporal-Variable Compositional Turning Decomposition Network (LTV-CTDNet), a
forecasting framework designed to combine competitive accuracy with structurally admissible outputs. LTV-CTDNet was evaluated using seven months of 15-minute
LiDAR observations from eight monitored corridor locations in Nashville, Tennessee. Its lightweight encoder combines recent turning-movement history, weekly time-slot
embeddings, and location
embeddings. The Compositional Turning Decomposition framework separately predicts nonnegative approach totals and within-approach turning proportions, then reconstructs movement forecasts from these components. Among the evaluated predefined configurations, LTV-CTDNet achieved a movement-level MAE of 1.8189 and RMSE of 3.8072. Its accuracy gains over the strongest sequence models were modest, but it produced no negative forecasts, while unconstrained learned models generated negative values in approximately 10.6% to 29.2% of raw forecast
cells. The framework enforces nonnegative outputs and exact agreement between each model-predicted approach total and the sum of its component movements by construction, providing directly interpretable forecasts without clipping or coherence correction.
Read the full abstract ▾
Algebraic Topology
A study of $n$-valued non-split maps and applications
Daciberg Lima Gonçalves, Bartira Maués and Daniel Vendrúscolo
For a given $n$-valued map, there exists a suitable minimal covering of the domain such that the composition of this covering with the map yields a split map, composed of $n$ single-valued maps, called the lift factors. Using the group action of the deck-transformation group we construct intermediate covering spaces as well as intermediate configuration spaces, allowing the original map to lift to these spaces, building a tower to illustrate at which level the lift factors split off. We then display two applications of these new tools. The first is to derive results concerning the Borsuk-Ulam property for certain coverings of the tower. The second is to obtain a sharp formula for the Nielsen number of an $n$-valued non-split map, using the coincidence number of specific single-valued maps that arise in the lifts of the tower.
Read the full abstract ▾
Machine Learning
When Known Physics Helps Neural PDE Models: Residual Constraints Out-Regularize Generic Priors for Nonlinear Dynamics
Zahra Farazpay and Aniruddha Bora
Neural PDE surrogates increasingly incorporate structural priors, yet it is often unclear whether their gains arise from physics-specific information or simply from regularization and training choices. We evaluate several such priors under a common protocol against a matched from-scratch neural operator baseline. Our central result is that a known-equation residual consistently outperforms the best generic regularizer at equal tuning budget. At fixed capacity this benefit appears across linear and nonlinear PDEs, but a capacity sweep reveals a sharp distinction: the advantage persists and grows for Burgers, KdV, and Allen-Cahn, while collapsing toward or below parity for linear heat and advection-
diffusion. Thus, the durable value of the residual is specific to nonlinear operators. We further falsify a pre-registered hypothesis that the benefit is activated only by data sparsity: the residual remains advantageous even under full supervision. Its usefulness does, however, have a clear boundary. Under grid under-resolution, nonlinear coarse fields no longer satisfy the naive governing-equation residual, and enforcing it becomes actively harmful. In contrast, cross-family pretraining and in-context conditioning fail to outperform the strong from-scratch baseline in the regime studied. Together, these results identify when known physics provides non-redundant information to neural PDE models, when it does not, and when enforcing it introduces bias.
Read the full abstract ▾
Operator Algebras
Selfless nonnuclear crossed products
Jamie Bell
For a strictly ergodic action of a countable discrete group with Ozawa's property PHP on the Cantor set, we prove that dynamical comparison implies complete selflessness of its reduced crossed product. Therefore these crossed products have stable rank one, strict comparison, a unique quasitracial state, and real rank zero. Applications include probability-measure-preserving Cantor actions of C*-simple acylindrically
hyperbolic and C*-simple linear groups for which an infinite amenable subgroup acts strictly ergodically. The results also apply to subodometers and infinite regular Toeplitz subshifts of residually finite groups with PHP.
Read the full abstract ▾
Robotics
ZeroBot: Learning from Scratch in Minutes with Generative Real2Sim
Ivan Kapelyukh, Xiaohan Zhang, Stephen James et al. (+2 more)
We present ZeroBot, a real2sim framework for learning a
robot manipulation task from scratch in minutes under challenging conditions: zero human demonstrations, zero policy pre-training, and zero known object models. Given only a single view of an object and a goal pose for that object, ZeroBot uses image-to-3D generative models to obtain a complete object mesh, which is used in simulation for large-scale parallel reinforcement learning. To accelerate training, we introduce an action space which leverages the generated geometry and learned value function to sample states involving
robot-object contact. When evaluated on real-world tasks including grasping, pushing, articulated object interaction, and multi-stage manipulation, ZeroBot achieves an 87% success rate with an average training time of 119 seconds. These results show the value of using image-to-3D models in a real2sim framework for rapid, autonomous
robot learning.
Read the full abstract ▾
All authors Ivan Kapelyukh, Xiaohan Zhang, Stephen James, Laura Herlant, Edward Johns
Machine Learning
Fisher-Informed Recalibration for Feedback-Based On-Policy Self-Distillation of LLMs
Seohyun Lee, Dong-Jun Han, Seyyedali Hosseinalipour et al. (+1 more)
Feedback-based on-policy self-
distillation has emerged as a promising approach for enabling foundation models, more specifically Large
Language Models (LLMs), to learn from their own outputs under external feedback, with a single model serving as both teacher and student. However, such methods can exhibit unstable optimization, conducive to performance collapse during training. To address this limitation, we propose FIRE (Fisher-Informed REcalibration), a dual-branch framework that recalibrates the supervision applied to correct and incorrect on-policy outputs during fine-tuning. For correct responses, FIRE replaces self-
distillation with re-weighted on-policy SFT, while for incorrect ones FIRE identifies feedback components that disproportionately influence the teacher-induced update and recalibrates the feedback-conditioned target accordingly. Both branches are influenced by a
token-level radius derived in part from a softmax Fisher trace. FIRE separates which direction feedback should move the model from how far the model should move in that direction, while leaving well-behaved feedback supervision unchanged. Our experiments demonstrate that FIRE provides substantially more stable self-
distillation while maintaining strong downstream performance, particularly in settings where standard feedback-conditioned
distillation becomes unstable.
Read the full abstract ▾
All authors Seohyun Lee, Dong-Jun Han, Seyyedali Hosseinalipour, Christopher G. Brinton
Materials Science
Correlation Between Dopant Atom Evaporation Field and Measured Site Preference in Atom Probe Tomography
Kasper A. Hunnestad, Constantinos Hatzoglou, Frida H. Danmo et al. (+6 more)
Complex
oxides possess a wide range of electric,
magnetic, and optical properties that can be precisely tuned by chemical doping. The atomic-scale analysis of the property-controlling dopants, however, becomes increasingly difficult towards low doping levels. Atom probe
tomography (APT) offers chemical sensitivity and spatial resolution to image individual dopant atoms down to a few parts per million. To reliably extract such information, detailed knowledge about the atom-specific field evaporation processes is required. Here we demonstrate a first insight into the APT-measured atomic position of dopant atoms and the field evaporation conditions, using Zr-doped ErMnO3 as a model system. Our analysis reveals a substantial preferential retention of both
matrix and dopant atoms which strongly affects the dopant site determination and can lead to an incorrect interpretation. The retention effect is determined by intrinsic and extrinsic parameters, such as the dopant's evaporation field and concentration and the analysis temperature, respectively, as we explain based on field-evaporation simulations. Our results are important for the APT-based analysis of individual dopant atoms in solid systems and the understanding of field evaporation dynamics at the atomic level in general.
Read the full abstract ▾
All authors Kasper A. Hunnestad, Constantinos Hatzoglou, Frida H. Danmo, Zewu Yan, Edith Bourret, Francois Vurpillot, Antonius T. J. van Helvoort, Sverre Selbach, Dennis Meier
Artificial Intelligence
EHRAdapt: Adapting Pretrained Language Models to Electronic Health Records with Semantic Priors for Rare Clinical Events
Andre R Goncalves, Vincent Liu and Priyadip Ray
Electronic health records (EHRs) encode clinical histories as (time, modality, code) tuples, whereas pretrained
language models expect text
tokens. Serializing tuples as text inflates sequence length and repeatedly encodes the same structure. We introduce EHRAdapt, an adapter that maps tuples directly into a frozen
language model's
embedding space. Modality and time gaps enter through a learned
embedding and
attention biases, and event codes receive dedicated vectors. Learning them is challenging because clinical vocabularies are long-tailed, leaving rare events with few observations. EHRAdapt therefore represents each event vector as the sum of a semantic prior and an evidence residual. The prior is a frozen, ontology-trained biomedical
embedding of the event's clinical description, mapped into the input space by a shared learned projection, so it supplies clinical meaning even when observations are scarce. The residual is a learned low-rank, event-specific correction that refines it as evidence accumulates. We perform continued pretraining on records from about 4 million patients with three frozen LLM backbones, OLMo2 1B, Llama3.2 1B, and OLMo2 7B, training only the adapter (0.1–0.6% of all parameters). The full adapter outperforms all ablations in held-out next-event prediction on every backbone. Removing the semantic pathway hurts rare events most, over ten times more than the most frequent ones, whereas removing the residual hurts overall prediction but improves it for the rarest events. On reportable infectious-disease and syndromic classification, EHRAdapt outperforms text-based LLM and count-based baselines and the same LLMs adapted to the same records through next-event prediction, and both pathways improve rare-disease discrimination. The two pathways therefore play complementary roles, visible only when results are broken down by event frequency.
Read the full abstract ▾
Robotics
TacGooseBumps (TacGB): Retrofitting Normal-Only Tactile Sensors with Shear Encoding for Learning Contact-Rich Manipulation
Wenjie Li, Binyu Yang, Yuxin Chen et al. (+2 more)
Contact-rich policies often fail because distinct physical states look alike yet require different actions. Cameras may not reveal whether a connector is aligned or fully seated, while many normal-only tactile sensors can miss the tangential interactions perpendicular to the grasping direction that distinguish these states. We ask whether a learning policy needs calibrated shear measurements, or only a repeatable observation that separates shear-dependent contact states. We introduce TacGooseBumps (TacGB), a passive domed film that mechanically encodes tangential loading as pattern changes in an existing sensor's pressure map. Tangential loading tilts each dome and redistributes pressure across its footprint; an end-to-end policy consumes the resulting maps without added electronics, force reconstruction, or taxel-level dome
alignment. Across four imitation-learning tasks and two data-collection pipelines, TacGB improves goal attainment, efficiency, and contact quality: insertion success increases by up to 36 percentage points, and successful insertions are completed faster, while fragile-object placement becomes gentler and drawing becomes more continuous and straight. Signal, stage-wise, failure-mode, and trajectory analyses link these gains to contact regimes in which task-relevant tangential interactions are poorly resolved by
vision and normal pressure alone. Together, these results show that shear need not be measured metrically to benefit
robot learning; it can instead be mechanically encoded without changing the underlying tactile sensor or the policy's pressure-map input format.
Read the full abstract ▾
All authors Wenjie Li, Binyu Yang, Yuxin Chen, Ambrose Wang, Masayoshi Tomizuka
Physics and Society
Learning to cooperate in a changing world: How caring about the future promotes cooperation across scales
Yuxin Geng, Xingru Chen, Xin Wang et al. (+4 more)
In social dilemmas, individuals need to forgo short-term temptations to achieve synergistic collective outcomes through cooperation. Previous work has examined mechanisms through which cooperation can evolve, including direct reciprocity, indirect reciprocity, environmental stochasticity, network reciprocity, and demographic stochasticity. These mechanisms have largely been studied under natural selection or social learning, where strategies with higher payoffs are more likely to spread. It is equally important to study cooperation under self-learning, where individuals adapt through their own experiences. Here, we focus on multi-
agent reinforcement learning and derive analytical conditions under which these mechanisms stabilize cooperation under learning dynamics across scales. We find that reinforcement learning can steer self-interested individuals toward cooperation when they value the future over short-term temptation. Our work provides a unified approach to identifying the determinants of learning to cooperate in a changing world, thereby paving the way for the advancement of cooperative artificial intelligence.
Read the full abstract ▾
All authors Yuxin Geng, Xingru Chen, Xin Wang, Hongwei Zheng, Longzhao Liu, Shaoting Tang, Feng Fu
Machine Learning
RICE-Alpha: Reliability-Informed Correction with Event Graphs for LLM-Agent Stock Forecasting
Tong Liu, Lanmiao Liu and Xiang Hu
Equity-relevant news evolves through temporally dependent corporate events, making historical information useful only when event continuity, information availability, and transition reliability are modeled. Existing LLM-based financial
agents incorporate historical evidence, yet they provide limited support for preserving issuer-specific chronology under point-in-time constraints and for identifying when historical transitions contribute information beyond the current forecast. We present RICE-Alpha (Reliability-Informed Correction with Event Graphs), a point-in-time stock-scoring framework that separates a history-aware multi-view Base Alpha from a reliability-calibrated residual correction derived from historical event continuation. A Multi-Tier
Memory Layer grounds news interpretation in temporally eligible issuer-specific history, while a Typed Event
Agent constructs event states whose successor relations are formed within issuers and pooled across firms only after valid local pairing. Matured transitions are calibrated by their empirical reliability, and the resulting graph signal is residualized against the Base Alpha and technical view to obtain the RICE Delta. On daily Nasdaq-100 and Hang Seng Index panels from 2024 to 2026, RICE-Alpha achieves the strongest results among the evaluated LLM-based
agents and momentum across four predictive and four portfolio-level metrics. Its ICIR more than doubles that of the strongest baseline, while net Sharpe ratios reach 1.656 and 1.725 in the U.S. and Hong Kong, respectively. U.S. ablations further show significant reductions in IC and RankIC after Holm adjustment when major components are removed. These results indicate that historical event continuation adds incremental information when it is temporally grounded, reliability-calibrated, and introduced as a residual correction to a multi-view forecast.
Read the full abstract ▾
Cryptography and Security
SweepMark: Query-Free and High-Fidelity Attacks on Robust Audio Watermarking
Weikang Ding, Binhao Ma, Hanqing Guo et al. (+1 more)
Audio watermarking embeds imperceptible signals into
speech to trace misuse in text-to-
speech (TTS) synthesis. Existing removal attacks may retain watermark-related artifacts or degrade
speech quality when suppressing them. We propose SweepMark, a query-free black-box attack that targets these artifacts without access to target watermarking models or paired clean and watermarked samples. Our key insight is that watermark
embedding introduces spectral patterns that are not fully aligned with natural
speech, including persistent structures that conventional
speech enhancement may retain. Diverse Artifact Learning extracts complementary stationary and non-stationary patterns, using customized spectral artifacts added to public
speech to learn persistent structures without target watermark examples. Adaptive Artifact Scaling then combines and amplifies these patterns for subtraction while constraining distortion. Across two
speech datasets and four watermarking methods, SweepMark achieves average attack success rates of 92% and 96%, with higher perceptual quality scores overall than the evaluated adaptive attacks. Automatic
speech recognition evaluation indicates that linguistic content is largely preserved. Further experiments with three TTS systems show high attack success against two synthesis-resistant watermarking methods in
speech generated after fine-tuning on attacked recordings.
Read the full abstract ▾
All authors Weikang Ding, Binhao Ma, Hanqing Guo, Rui Duan
Machine Learning
T-SNN: Temporal Simplicial Neural Network for EEG Decoding
Nikita Malik, Shubhajit Roy, Mohit Kataria et al. (+4 more)
Decoding
brain states requires models that capture both the
evolution of neural activity and interactions among groups of
brain regions. Existing
EEG methods often treat recordings as multivariate time series or represent functional connectivity with pairwise graphs, leaving dynamic higher-order interactions largely unmodeled. We introduce the Temporal Simplicial Neural Network (T-SNN), which represents
EEG recordings as sequences of evolving simplicial complexes. By combining simplicial convolutions with recurrent updates, T-SNN jointly learns higher-order interactions and their temporal
evolution. On the seven-class SEED-VII
emotion recognition task, T-SNN outperforms convolutional, recurrent, graph-based, and
Transformer methods in both trial-wise and cross-subject evaluations. Incorporating eye-movement features further improves performance, demonstrating the framework's potential for
multimodal brain-state decoding.
Read the full abstract ▾
All authors Nikita Malik, Shubhajit Roy, Mohit Kataria, Isuru Herath, Suraj Yadav, Inés García-Redondo, Dhananjay Bhaskar
Biomolecules
Assay-Aware BindingDB: Curating Experimental Context for Binding Affinity Prediction
Ming-Hsiu Wu, Xuejiao Shirley Guo, Bingsong Zeng et al. (+5 more)
Protein–ligand binding affinity prediction is fundamental to computational drug discovery, yet modern AI-driven models are limited by pervasive heterogeneity in their training data: bioactivity values are aggregated across diverse assay types and experimental conditions without accounting for protocol-level differences, introducing systematic noise. Existing harmonization approaches either discard assay-level metadata or collapse it into coarse categorical distinctions, leaving rich contextual signal unused. We address this gap with two contributions. First, we introduce Assay-Aware BindingDB, augmenting 74,425 BindingDB
protein–ligand pairs across four assay types (ITC, SPR, RBA, and FPA) with structured metadata extracted from primary literature using a two-stage agentic framework that separates evidence extraction from ontology-conditioned JSON synthesis. Against domain-expert references, the framework achieves presence F1 $\geq 0.947$ and semantic content accuracy $\geq 0.913$ across all assay types. Second, we develop a context-conditioned affinity model that injects an assay-context
embedding into the Boltz-2 affinity module. On a paper-level held-out split, the model reduces combined-regime MSE from $1.27$ to $1.19$ and increases Pearson correlation from $0.64$ to $0.67$, with significant gains for SPR and FPA. ITC, the only label- and immobilization-free assay considered, shows no improvement, consistent with its design removing the protocol artifacts the metadata captures. These results support the hypothesis that systematically curated assay metadata provides informative signal for affinity prediction.
Read the full abstract ▾
All authors Ming-Hsiu Wu, Xuejiao Shirley Guo, Bingsong Zeng, Ziqian Xie, Shuiwang Ji, Wenshe Liu, Cui Tao, Degui Zhi
Mesoscale and Nanoscale Physics
Topological chiral edge modes in the continuum of a trivial bulk
Li Ge
Chiral edge modes in topological states of matter are routinely found inside band gaps, which may be deformed and shifted as a function of the
lattice momentum. Here we show that they can exist inside a bulk band instead, i.e., in the continuum. These
chiral edge modes in the continuum (CEMICs) are induced by introducing an imaginary
magnetic flux to an otherwise topologically trivial two-dimensional system, realized by imposing asymmetric couplings along the edge. While the stronger couplings provide a preferred circulation direction, we find that the edge localization of CEMICs does not originate from the non-Hermitian skin effect, which is typically associated with systems featuring asymmetric couplings and the open boundary condition. Instead, this edge localization is due to the energy exchange between the system and the environment at the asymmetric coupling junctions, which leads to a parity-time (PT) transition in the Brillouin zone. Intriguingly, the onset of CEMICs is independent of the system size in the macroscopic limit, which is given by the golden ratio between the asymmetric couplings and their
geometric average.
Read the full abstract ▾
Audio and Speech Processing
Rethinking Automated Voice Similarity by Shifting from EER to Embedding Geometry
Szu-Chi Chen, Jia-Kai Dong, Yi-Cheng Lin et al. (+2 more)
Speaker verification (SV) models are commonly assumed to better capture nuances among speaker characteristics as verification accuracy improves, leading to their widespread use as automated proxies for human voice similarity in
speech generation tasks. However, by establishing a human perceptual
alignment metric and conducting systematic analysis, we demonstrate that perceptual
alignment is governed far more by how a model is trained (its learning objective) than by how well it performs (EER). Notably, standard margin-based classification losses (e.g., AAM-Softmax) yield substantially lower perceptual
alignment than prototypical metric losses, while EER itself fails to track human judgment, directly challenging the community's implicit assumption. We trace this divergence to
embedding geometry, where a model's effective dimensionality ($d_{\mathrm{eff}}$) tracks perceptual
alignment with a $-0.95$ rank correlation, revealing that the dimensional spread favored by classification losses fundamentally clashes with the low-dimensional nature of human voice perception. Imposing a dimensionality bottleneck compresses $d_{\mathrm{eff}}$ and raises perceptual
alignment ($\rho_{\mathrm{align}}$) from 0.08 to 0.74, establishing a principled
geometric criterion for evaluating voice similarity.
Read the full abstract ▾
All authors Szu-Chi Chen, Jia-Kai Dong, Yi-Cheng Lin, Sung-Feng Huang, Hung-yi Lee
Computer Vision
MetaSampling: Making Frame Samplers Efficient for Long-Video Question Answering
Ashim Dahal and Bikramjit Banerjee
Frame selection is an important component of long-
video question answering (VQA) with
Multimodal Large
Language Models (MLLMs). Existing frame-selection methods improve over simple top-$k$
embedding retrieval and uniform sampling, but are typically applied under a fixed global selection budget. We introduce MetaSampling, a training-free, plug-and-play sampling strategy that can be applied on top of existing frame selectors. MetaSampling improves downstream VQA efficiency by dynamically reducing the number of frames passed to the MLLM while preserving, and in some cases improving, answer accuracy. We evaluate MetaSampling across 36 paired frame-selector–MLLM-backbone–VQA-
benchmark configurations. MetaSampling reduces the number of selected frames in all 36 configurations and improves accuracy in 25 of them, yielding an average frame reduction of $8.9\%$ while slightly improving accuracy overall.
Read the full abstract ▾
Quantum Physics
QBX: A Compiler for 2-local Qubit Hamiltonian Simulation on Quantum Chiplets
Zikun Li, Zhuoming Chen and Zhihao Jia
2-local
qubit Hamiltonian simulation, a fundamental task in
quantum computing, is widely applied in various applications. This paper presents QBX, the first
quantum compiler designed for 2-local
qubit Hamiltonian simulation on
quantum chiplet architectures. Existing general-purpose
quantum compilers for chiplet architectures target
quantum programs at the gate level and miss optimizations requiring high-level semantics of
Hamiltonian simulation. Conversely, existing domain-specific compilers for
Hamiltonian simulation are designed for monolithic architectures, and their limited scalability and inability to adapt to the heterogeneity of chiplet architectures lead to sub-optimal
circuit outputs. QBX implements a scalable hierarchical approach to reduce both the amount and distance of cross-chiplet communications. It integrates the highway mechanism, a chiplet-oriented solution for efficient long-range
qubit communication, to diminish cross-chiplet interaction costs. Furthermore, QBX efficiently aggregates 2-local Pauli strings into multi-target controlled gates, harnessing highways for deep optimization. Our evaluations show that QBX markedly outperforms both general-purpose and domain-specific compilers in
circuit depth and operation count, while scaling to much larger 2-local
Hamiltonian simulations.
Read the full abstract ▾
Computer Vision
UnfoldCRF: Structured Mask Refinement with Image-Conditioned Latent Regions
Chunming He, Rihan Zhang, Lei Xu et al. (+5 more)
Learned mask refiners improve
segmentation accuracy, but it is hard to tell how much of the improvement comes from explicit structure rather than from extra capacity, and whether it holds up when the mask generator or its error distribution changes. UnfoldCRF treats refinement as inference in a conditional random field over pixel labels and latent region variables. Its energy has a corrected unary term, learned local pairwise interactions, and image-conditioned latent-region consistency, with a null state that lets a region with weak label agreement withdraw from the consistency term; inference unrolls damped mean-field updates on this one energy. To isolate the effect of structure, we compare against recurrent black-box refiners that read the same inputs and receive the same parameter budget, stage count, and supervision. On COD10K, UnfoldCRF beats the strongest matched control by 1.0 $F^\omega_\beta$ point, improves all four COD metrics, and lowers the fraction of images made worse from 11.7% to 8.5%. Under a train-once protocol over five
datasets and several mask sources, the 2.6M-parameter variant gains 4.2 mean $\Delta$IoU against 2.0 for its control, and a variant built on frozen DINOv2 features matches the strongest foundation-model refiner with about a seventh of its resident parameters while staying ahead of its own control. On mask generators never seen in training, the gain is 2.0 $F^\omega_\beta$ points against 0.6 for the control. Zeroing individual messages shows where the corrections come from: the pairwise messages mostly fix boundaries, the region messages mostly fix non-boundary errors. Code and supporting materials will be publicly released.
Read the full abstract ▾
All authors Chunming He, Rihan Zhang, Lei Xu, Guanyi Qin, Chengyu Fang, Longxiang Tang, Fengyang Xiao, Sina Farsiu
Differential Geometry
Codimension-four regularity of noncollapsed Ricci limit spaces under an integral volume-deficit bound
Lingling Kong
Let $(X,d,p)$ be an $n$-dimensional noncollapsed
Ricci limit space, where $n\ge4$. Under an integral volume-deficit bound, we prove that the metric singular set has Hausdorff codimension at least four and sigma-finite $(n-4)$-dimensional Hausdorff measure. This establishes a special case of the codimension-four regularity
conjecture. Moreover, the nonmanifold locus is closed and has locally finite $(n-4)$-dimensional Hausdorff measure, and its intersection with each bounded ball satisfies a tubular-volume estimate of order $r^4$. In dimension four, the nonmanifold points form a locally finite set. A flat quotient example shows that the codimension bound is sharp.
Read the full abstract ▾
Plasma Physics
A packet-level digital hardware twin for commissioning megahertz diagnostic edge AI and plasma control system integration in tokamaks
Semin Joung, Abhilasha Dave, Luca Scomparin et al. (+6 more)
High-bandwidth
plasma diagnostics increasingly provide inputs to machine learning and signal-processing
algorithms intended for real-time tokamak control, but the complete path from diagnostic sampling to control-system handoff is difficult to commission because of their sampling rates. We develop a packet-level digital hardware twin for the megahertz diagnostic edge-AI architecture. The simulator represents a 64-channel, 1 MHz beam emission spectroscopy (BES) diagnostic embedded in a 96-channel dual-carrier acquisition system, two 48-channel streams with SPAD0 sample counters, 10 GbE Hardware UDP transport, packet loss and network jitter, FPGA parsing and dual-carrier
alignment, causal preprocessing and edge inference, a compact Ethernet result packet, receiver-side shared state, and a 1 kHz PCS-like control cycle. Binary UDP payloads and PCAP files are generated rather than emulating transport only at the array level. With a baseline of 20 samples per packet, each carrier generates 50,000 packets s$^{-1}$ and 100 MB s$^{-1}$ of user payload. A 110 ms reference run produces 11,000 HUDP packets; an intentionally dropped 20-sample packet is detected by the sample-counter continuity logic and invalidates the two overlapping 128-sample inference windows without silent interpolation. For valid windows, the configured engineering latency model gives a median last-input-to-shared-
memory latency of 91.6 us and a 99th percentile of 108.1 us. A separate operating-system loopback test sends binary FPGA-result datagrams through a UDP receiver into POSIX shared
memory and preserves packet sequence and CRC for 20/20 packets. Interactive GUI interfaces expose timing, packetization, network faults, inference thresholds, and control-state inspection. The framework provides a reproducible environment for testing diagnostic-to-accelerator interfaces and fail-safe behavior for deployment on fusion devices.
Read the full abstract ▾
All authors Semin Joung, Abhilasha Dave, Luca Scomparin, Filipp Khabanov, Zheng Yan, Benedikt Geiger, George McKee, Ryan N. Coffee, David R. Smith
Machine Learning
ASTRA: ADMM-Accelerated Topology Reconfiguration for Dynamic Satellite Constellations
João Norberto, Ricardo Ferreira and Cláudia Soares
Dynamic
topology reconfiguration is central to the reliability and efficiency of large satellite constellations, yet many existing approaches rely on idealized assumptions such as full constellation deployment or uniform orbital spacing. We present Adaptive Satellite
Topology via Regret-Aware learning (ASTRA), a theoretically grounded framework for dynamic satellite
topology reconfiguration that builds on an online learning formulation and makes it computationally practical. ASTRA combines an ADMM-based offline solver with efficient online updates for both online gradient descent and online conditional gradient, yielding markedly cheaper constrained updates than generic optimization pipelines. On the theory side, we show that for a relevant class of entry-wise nonzero utility
matrices, the objective is strongly convex, which yields logarithmic static regret for online gradient descent, and we further instantiate known dynamic-regret guarantees under inexact ADMM inner loops. Empirically, ASTRA matches or improves
topology quality, presenting a good trade-off with computational time on synthetic constellations, and it remains effective on real Starlink data under partial deployment and non-uniform spacing, where idealized structural assumptions break down. These results position ASTRA as an efficient and theoretically grounded approach to
topology reconfiguration in realistic Low Earth Orbit networks.
Read the full abstract ▾
Cryptography and Security
GateDrain: Availability Attacks and Admission-Side Defense for Confidence-Gated Edge-Cloud Inference
Zonghua Gu, Julian Singh-Smith, Junlin Liao et al. (+1 more)
Confidence-gated edge–cloud inference accepts confident local predictions and offloads uncertain inputs to a stronger cloud model. We show that this routing decision creates an availability attack surface. We call this attack GateDrain: bounded input perturbations lower calibrated confidence and redirect requests that would otherwise be answered locally into a shared cloud queue, without increasing the application request rate. Because escalated requests share a cloud service, an increase in per-request cloud demand can move a near-capacity deployment across a queueing knee, causing disproportionate tail-latency degradation for benign users. We evaluate white-box, transfer, decision-only, universal, and multi-gate attacks on the public EdgeBoost artifact. A fixed-application-volume comparison isolates the effect of confidence manipulation from added client traffic, while perturbation-budget and arrival-process sweeps show that the queueing transition persists across several workload models but its amplification depends on the operating point. Adaptive attacks also defeat the evaluated training-free preprocessing defenses. To contain the resulting cloud demand, we evaluate Bounded Escalation, which combines per-source admission budgets, protected capacity, and non-preemptive trusted-class priority; an optional global bucket adds an identity-independent bound on untrusted admissions. The evaluation makes the resulting policy trade-off explicit: authenticated clients receive latency isolation, whereas tighter aggregate containment can reject legitimate unauthenticated offloads and reduce overall expected accuracy through edge fallback.
Read the full abstract ▾
All authors Zonghua Gu, Julian Singh-Smith, Junlin Liao, Di Liu
Computer Vision
A Multi-Dataset Benchmark of YOLO-Based Weed Detection in Precision Agriculture
Hristina Zdraveska, Vlatko Spasev, Ivica Dimitrovski et al. (+2 more)
Weed detection is an important component of precision agriculture, enabling site-specific weed management and reducing unnecessary herbicide use. Although deep learning methods have achieved strong results for crop and weed detection, many studies rely on single-
dataset evaluation, making it difficult to assess robustness across different agricultural domains. This paper presents a multi-
dataset benchmark of deep object detectors for weed detection in precision agriculture, with a focused evaluation of YOLO26 models. We evaluate nano, small, and medium variants on seven public weed-detection
datasets covering different crops, weed species, field conditions, acquisition setups, and annotation protocols. The models are compared in terms of detection accuracy, model complexity, inference latency, FPS, and model size. In addition to in-
dataset evaluation, we investigate cross-domain generalization using a unified one-class weed setup and evaluate multi-source training using the combined training subsets from all
datasets. The results show that YOLO26 achieves strong in-
dataset performance, with YOLO26m obtaining the highest average accuracy and YOLO26s providing the best practical accuracy-efficiency trade-off. However, cross-domain performance decreases substantially, with YOLO26s dropping from an average in-domain mAP$_{50:95}$ of 0.603 to 0.148 in the off-domain setting. Multi-source training improves performance on several
datasets, but does not fully eliminate domain shift. Overall, the
benchmark highlights the importance of
dataset diversity, domain similarity, and target-domain adaptation for robust weed detection in real-world precision agriculture applications.
Read the full abstract ▾
All authors Hristina Zdraveska, Vlatko Spasev, Ivica Dimitrovski, Ivan Kitanovski, Petre Lameski
High Energy Physics (Experiment)
First measurement of the antihydrogen production cross section through the charge-exchange reaction of low-energy antiprotons with orthopositronium
GBAR Collaboration: P. Adrich, I. Belosevic, M. Chung et al. (+41 more)
The GBAR experiment has measured the formation rate of antihydrogen from antiproton impact on a positronium cloud for antiproton kinetic energies of 4 and 6 keV. This is the first charge-exchange cross section measurement performed using antiproton beams, which are provided by the AD-ELENA facility at CERN. The measured cross section values are $(14.1 \pm 1.3 \mathrm{(stat)} ^{+2.2}_{-1.4} \mathrm{(sys)}) \times 10^{-16}$~cm$^2$ at 6.2 keV energy and $(8.7 \pm 2.4 \mathrm{(stat)} ^{+1.5}_{-0.09} \mathrm{(sys)}) \times 10^{-16}$~cm$^2$ at 4.15 keV energy, and agree with recent theoretical three-body calculations for antihydrogen formation. These measurements are an important input for experiments with antimatter, e.g the planned measurements of antihydrogen gravitational acceleration in the GBAR collaboration.
Read the full abstract ▾
All authors GBAR Collaboration: P. Adrich, I. Belosevic, M. Chung, P. Cladé, P. Comini, P. Crivelli, P. Debu, A. Douillet, S. Geffroy, S. Guellati-Khelifa, P. Guichard, P.-A. Hervieux, L. Hilico, P. Indelicato, S. Jonsell, J.-P. Karr, B. Kim, S. Kim, E.-S. Kim, N. Kuroda, B. Lee, L. Liszkay, D. Lunney, G. Manfredi, B. Mansoulié, V. Martimort, M. Matusiak, V. Nesvizhevsky, F. Nez, K. Park, N. Paul, E. Perez, P. Pérez, C. Regenfus, C. Roumegou, J.-Y. Roussé, F. Schmidt-Kaler, K. Szymczyk, T. A. Tanaka, B. Tuchming, D. P. van der Werf, D. Won, S. Wronka, P. Yzombard
Computation and Language
RewardExplainer: Learning Reward Model Explanations from Counterfactual Preference Feedback
Jingyi He, Nier Wu, Shuang Liu et al. (+3 more)
Reward models (RMs) are a key component of large
language model post-training, providing reward signals for subsequent reinforcement learning. However, conventional discriminative RMs typically output only scalar scores, making it difficult to identify the response behaviors associated with their scoring decisions. Existing interpretation methods often rely on predefined high-level attributes and require repeated counterfactual interventions for each response pair to validate candidate explanations, lacking a closed-loop mechanism that uses RMs' feedback to train a reusable explainer. To address this, we propose RewardExplainer, a framework that obtains feedback from the target reward model through counterfactual rewriting and uses this feedback to further optimize the explainer. RewardExplainer generates open-ended, atomic, and intervenable natural-
language scoring mechanisms, making explanations more concrete, readable, and actionable. It further converts counterfactual feedback into preference supervision, enabling the explainer to more faithfully capture the target RM's scoring preferences and sensitive behaviors than single-pass generation. Extensive experiments across multiple target RMs and explainer backbones show consistent improvements. Beyond interpretation, we use the generated mechanisms to identify potential bias patterns and construct targeted debiasing data for fine-tuning the reward model, improving robustness on reward-hacking
benchmarks.
Read the full abstract ▾
All authors Jingyi He, Nier Wu, Shuang Liu, Xin Wang, Mengnan Du, Xia Hu
Systems and Control
Regime-Dependent Value of CVaR in Preventive Maintenance Scheduling under RUL Uncertainty
Jerzy Baranowski and Waldemar Bauer
We study preventive-maintenance scheduling for a small fleet when several assets compete for a limited number of maintenance slots and their remaining useful lives (RULs) are uncertain. The planning horizon is divided into discrete periods; each asset may be maintained at most once, and at most $K$ assets may be maintained in any one period. Future usage and RUL errors are represented by scenarios. We compare a risk-neutral policy that minimizes expected cost with a risk-aware policy that minimizes expected cost plus conditional value-at-risk (CVaR), where $\operatorname{CVaR}_{0.90}$ is the mean cost among the worst 10% of scenarios. Both policies are solved exactly by enumerating all capacity-feasible schedules. A $3\times5$ experiment combines capacities $K=1,2,3$ with five increasing levels of RUL uncertainty, with ten paired replications for each combination. The risk-aware policy provides its largest benefit when RUL uncertainty is low-to-moderate and more than one maintenance action can be executed per period. In the best observed case, it reduces CVaR by 13.4% with an expected-cost increase of only 0.2 relative cost units. At high uncertainty, the benefit becomes negligible and the two policies can select identical schedules. The results show that risk-aware scheduling creates operational value only when prognostic information is sufficiently informative and the maintenance system has enough flexibility to act on it.
Read the full abstract ▾
Computation and Language
Opera: A Verbal Critic Framework for Long-horizon Coding Agents
Kai Mei, Zhiyuan Hu, Yutong Dai et al. (+7 more)
Long-horizon coding
agents need timely corrections, yet feedback can be ineffective or even harmful when it misjudges ongoing work or fails to address the underlying problem. Existing critics focus on evaluating trajectories and generating feedback, but rarely track what happens after feedback is delivered. We present Opera, a verbal critic framework that treats each correction as a persistent note, followed until the diagnosed problem is resolved. Opera decides when to review through periodic and event-driven triggers, diagnoses issues with typed operators, audits feedback against visible evidence before delivery, and tracks the
agent's subsequent actions to distinguish mere compliance from actual resolution. As a test-time critic, Opera improves the resolve rate of non-critic
agents by up to 12.4, 15.0, and 8.9 percentage points on Terminal-Bench 2.1, a SWE-Bench Pro subset, and DeepSWE v1.1, respectively, across four policy models, and achieves the highest mean resolve rate among competitive critic baselines on all three
benchmarks, and also improves policy models when the policy critiques itself. Beyond inference, Opera-guided rollouts provide approximately on-policy training data: fine-tuning Qwen3.5-9B on them improves its resolve rate on held-out SWE-Bench Pro repositories by 10.2 percentage points without a critic at inference time, matching fine-tuning on rollouts from a stronger model, while preserving its performance when switching harness, i.e., from Openhands to Terminus-2, which the latter substantially degrades. Our code is available at: https://github.com/dongyuanjushi/Opera.
Read the full abstract ▾
All authors Kai Mei, Zhiyuan Hu, Yutong Dai, Juntao Tan, Yifan Zhang, Dingjie Song, Dimitris N. Metaxas, Silvio Savarese, Ran Xu, Zeyuan Chen