Peer Revenue Publish or profit

Papers

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

All of arXiv

Friday 4 September

Computer Vision

Lightweight Vision Transformer Compression for On-Device Plant Disease Detection in Resource-Constrained Agricultural Field Conditions

Chilli (Capsicum annuum) is one of India's most economically significant crops, yet its productivity is persistently threatened by diseases that are difficult to identify without expert intervention. While Vision Transformers (ViTs) have achieved high classification accuracy, their large computational footprint makes deployment on resource constrained devices challenging. Existing compression approaches typically address pruning, quantization, and knowledge distillation in isolation, leaving the potential benefits and interactions of their combined application insufficiently explored. We propose a unified Vision Transformer compression framework that combines Hessian-Balanced Adaptive Block Pruning (H-BAC), guided by second-order sensitivity estimation, with quantization and attention-based knowledge distillation. To systematically identify the most effective configuration within each compression family, each technique is first evaluated independently through controlled ablation studies, after which the best-performing components are integrated into a sequential deployment pipeline tailored to real-world agricultural constraints. On a chilli 3-class village-split dataset with a genuine cross-village, cross-device out-of-distribution test split, the resulting compressed models match or exceed the 95.13% FP32 baseline's accuracy, alongside 74-98% model size reduction, and the fully integrated compression pipeline achieves a 54.5x size reduction (327.42 MB to 6.01 MB) at 95.13 +/- 2.32% accuracy across four tested configurations. A direct comparison further reveals that, on this dataset, a directly-trained student of the same final size, without pruning or distillation, reaches comparable accuracy of 94.87%, at the same 6.01 MB INT8 size, indicating where H-BAC and knowledge distillation are, and are not yet shown to be, worth their computational cost.
Read the full abstract

All authors Mahadev Sunil Kumar, Bhavika Gondi, Desaisetty Venkata Satya Sai Swapnith, Gangireddy Rahul Jogi, Sudheesh Manalil, Arnab Raha, Amitava Mukherjee, Parthasarathy Seethapathy, G. Gopakumar

Artificial Intelligence

Technical Manual for a Toolkit for Measuring Contextual Individuation in Transformer Language Models

A transformer language model assigns a single, context-independent vector to a word type at its embedding layer, yet is widely believed to individuate that word's occurrences by context in its later layers. Testing this belief cleanly requires a construct that holds the word form fixed while its context and intended sense vary in a controlled, labeled way. This manual documents an open toolkit built around such a construct, which we call a bridge form: a single written word that recurs, unchanged, across two or more subject domains with a different sense in each. We describe, and justify, every stage of the pipeline: the declarative specification of bridge forms and their source domains, corpus acquisition from Wikipedia, occurrence localization, layer-wise representation extraction, a domain-pairwise silhouette measurement of separation in the model's representation space, and a paired visualization protocol. Each design choice is presented together with the methodological failure mode it is meant to avoid (sense contamination from overly broad category labels, the multi-group bias of the silhouette coefficient, subword-tokenization misalignment, and axis-comparability artifacts in dimensionality-reduced plots, among others). This manuscript is a methodological and implementation reference: it does not report or interpret empirical outcomes of running the toolkit on any particular model or bridge-form set. The toolkit, its full source, and the corpora used to exercise it are archived separately (Section 9) under a persistent identifier, and are intended to be cited as an instrument by studies that use it to produce and interpret empirical results.
Read the full abstract

All authors José Luciano Verçosa Marques, Frederico Jorge Heitmann, Daniel Omar Perez, Marcelo Vinicius de Paula, Tárcio André dos Santos Barros

Systems and Control

Headroom-Aware Stochastic Adaptive Model Predictive Control for Load Frequency Control in Microgrids

As the penetration of inverter-based resources (IBRs) increases in microgrids, they are increasingly expected to play a greater role in load frequency control (LFC). Model predictive control (MPC) is attractive for LFC because it incorporates system dynamics and operational constraints. However, most MPC-based LFC formulations rely on fixed reserve headroom based on forecasted renewable availability or storage systems. Under short-term renewable intermittency, IBR headroom is stochastic and time-varying, causing optimal control commands to exceed the physically deliverable regulation capability and cause stochastic saturation. This control-actuator mismatch degrades LFC performance. Accordingly, this study develops two headroom-aware strategies for PV-dominated microgrids. First, stochastic headroom constrained MPC (SHCMPC) incorporates headroom predictions through time-varying input constraints to enforce control feasibility. Second, stochastic adaptive MPC (SAMPC) embeds headroom awareness into the MPC objective function by adaptively penalizing control actions based on predicted headroom, reducing reliance on units with limited headroom without hard time-varying constraints. Simulation results show that stochastic saturation degrades conventional MPC-based LFC, particularly under tight reserve margins. Both strategies improve regulation performance. SHCMPC eliminates saturation events, while SAMPC achieves substantial saturation mitigation with lower computational effort, offering a computationally efficient alternative for real-time LFC under stochastic renewable availability.
Read the full abstract

All authors Erfan Mehdipour Abadi, Shuo Yuan, Le Yi Wang, Caisheng Wang, Feng Lin

Robotics

Adaptation Needs in Robotic Systems: Assessing Behavior Trees and Their Enhancement

Robotic systems increasingly operate in dynamic, uncertain, and open-ended environments, where design-time assumptions may no longer hold, and adaptation becomes necessary to maintain effective and safe operation. Behavior Trees (BTs) are widely used in robotic control architectures due to their modularity, readability, and reactivity. This raises a central question: are BTs sufficient to meet the adaptation needs of modern robotic systems? This paper investigates this question through a literature-driven study complemented by empirical validation. First, we derive a classification of robotic adaptation needs from the literature, organizing them into six categories: Knowledge, Perception, Actuation, System, Mission, and Environment. Then, we analyze the capabilities and limitations of classical BTs with respect to these needs. Then, we characterize BT-based approaches for adaptation from the existing literature and organize them into four primary families, i.e., generation, extension, evolution, and refinement, including approaches that combine multiple families. Our analysis shows that the modularity, flexibility, and reactivity of classical BTs are insufficient for adaptation needs involving runtime restructuring, reasoning under uncertainty, mission reinterpretation, learning, or integration with external knowledge and planning mechanisms. Enhanced BT approaches address several of these limitations, but to different extents and often with limitations of their own. Our findings relate adaptation needs to both the capabilities and limitations of classical and enhanced BTs, providing guidance on when classical BTs are sufficient, when enhanced mechanisms are needed, and which challenges remain or emerge for adaptive robotic control architectures.
Read the full abstract

All authors Mehran Rostamnia, Gianluca Filippone, Ricardo Caldas, Patrizio Pelliccione

Statistics Theory

Joint Distributions of Minimum and Maximum Angles on High-Dimensional Spheres

Consider $n$ independent random vectors sampled from uniform distribution on $(p-1)$-dimensional unit sphere. This paper investigates the limiting joint distribution for the minimum and the maximum values of their pairwise angles. It proves that the minimum and the maximum angles are asymptotically independent when both $n$ and $p$ tend to infinity, which solves an open problem raised in Cai, Fan and Jiang (2013) [Journal of Machine Learning Research 14, 1837-1864]. Cai, Fan and Jiang (2013) obtained the limiting marginal distributions for both the minimum and the maximum angles under assumption $\lim_{n\to\infty}{\ln n}/p=\beta$ according to whether $\beta=0$, $\beta\in (0,\infty)$, or $\beta=\infty$. This paper presents unified limits for both joint distributions and marginal distributions regardless of the relative divergence rate of $n$ and $p$. The paper also derives the limiting distributions for some statistics based on the minimum and the maximum angles,
Read the full abstract

Cryptography and Security

Machine Unlearning as Private Retroactive Algorithms

Machine unlearning typically aims to emulate retraining from scratch: upon a deletion request, the unlearning algorithm should produce an outcome that would have been obtained had the deleted point never been included. Recent work has shown that this emulation requirement carries no meaningful privacy semantics against an adversary who observes a sequence of releases. Machine unlearning is thus not a privacy question per se, but rather a data maintenance question, which is precisely the subject of retroactive algorithms. These are algorithms supporting modifications of past operations, guaranteeing that all subsequent answers reflect the revised history as if it had always been in force. We put forward a definition of private retroactive algorithms, combining the retroactivity requirement with differential privacy under continual observation. We present constructions achieving both privacy and retroactivity at no asymptotic cost over privacy alone for linear statistics, clustering, and histograms, alongside impossibility results.
Read the full abstract

All authors Haim Kaplan, Refael Kohen, Yishay Mansour, Kobbi Nissim, Uri Stemmer

Machine Learning

Embedded Graph Flows for Categorical Graph Generation

Generating categorical graphs requires choosing node and edge types that form a coherent structure without depending on node order. Many graph generators encode categories as fixed one-hot vectors, which can impose an artificial geometry in which categories are equidistant. We propose Embedded Graph Flows (EGF), a generative model that learns continuous embeddings for node and unordered-edge categories and transports Gaussian noise towards these learnt endpoints using a permutation-equivariant graph transformer. A terminal readout maps the embeddings back to discrete graph categories. Across molecular benchmarks, EGF achieved competitive performance. On QM9, EGF gives the best result on all four reported metrics among the three methods, including a Fréchet ChemNet Distance (FCD) of 0.150, compared with 0.717 for the categorical-diffusion baseline DiGress and 0.812 for the bridge-based baseline GruM. When applied to larger molecules in ZINC250k, EGF retains the lowest maximum mean discrepancy (MMD) using the neighbourhood subgraph pairwise distance kernel (NSPDK), indicating close agreement with the local substructures of the reference molecules. Our code is available at https://github.com/Trusted-System-Lab/EGF.
Read the full abstract

All authors Ethan Ma, Zihan Wang, Chris Siu Yeung Chow, Xinguo Feng, Qingqing Li, Rui Jiang, Naipeng Dong, Guangdong Bai

Artificial Intelligence

LLM-Driven Algorithm Design for Quantum Circuit Synthesis based on Binary Decision Diagrams

Quantum circuits are central to implementing quantum algorithms on quantum devices, where quantum gates must be reversible. Many quantum algorithms rely on Boolean functions, which must therefore be implemented reversibly within quantum circuits. Reversible circuit synthesis provides a way to translate such Boolean functions into reversible circuits. Binary decision diagrams (BDDs) offer a scalable approach to this task, but the resulting BDDs and circuits depend heavily on variable ordering. Existing ordering heuristics commonly minimize BDD size because it is closely tied to the circuit size. However, BDD size is an imperfect proxy for the quantum cost of the synthesized circuit (QCC). We propose QuantumEvo, an evolutionary framework that uses an LLM as a heuristic generator for QCC-aware BDD variable ordering. Instead of predicting orderings directly, QuantumEvo searches over ordering heuristics initialized from multiple heuristic families. Candidate heuristics directly manipulate variable orderings using standard BDD operations and are selected by downstream QCC. The discovered heuristic, HGA-QE, modifies the sifting step inside a genetic algorithm so that the procedure is better aligned with QCC. Across the benchmark set, HGA-QE achieves a 70.9% tie-or-win rate against the per-function best baseline and is strictly best on 13.5% of the functions. The results demonstrate broadly competitive QCC performance, with HGA-QE showing a clearer relative advantage in strict wins on the two benchmark suites drawn from sources different from the data used for heuristic discovery.
Read the full abstract

All authors Yoonju Sim, Federico Berto, Chuanbo Hua, Jinkyoo Park, Changhyun Kwon

Earth and Planetary Astrophysics

Contrasting C/O ratios in Uranus and Neptune from disequilibrium chemistry: A clue to distinct evolutionary pathways?

The formation of Uranus and Neptune remains poorly constrained largely due to uncertain deep elemental abundances. Carbon monoxide (CO), a disequilibrium species in the upper troposphere, provides an indirect constraint on the deep oxygen abundance. We investigate the deep O/H and C/O ratios of the ice giants and their formation history by accounting for meridional variations in atmospheric structure and uncertainties in chemical kinetics. We extended a 1D thermochemical and diffusion model into a pseudo-2D model by including latitudinal variations in key model parameters. The O/H ratio was inferred by matching the modeled upper-tropospheric CO mole fractions to observations, and combined with the deep carbon abundance to derive the C/O ratio. For Uranus, varying tropospheric methane alone yielded O/H $\sim [47-57] \times$ protosolar, whereas allowing $K_{zz}$ to vary expanded the range to O/H $\sim [62-177] \times$ protosolar. For Neptune, the corresponding ranges are O/H $\sim [182-215] \times$ protosolar and O/H $\sim [222-342] \times$ protosolar, respectively, supporting lower oxygen enrichment in Uranus than Neptune. Chemical-network uncertainties have a more modest effect, amounting to at most $\sim 10\%$ of the retrieved oxygen ranges, comparable to the uncertainty associated with the measured CO abundance on Neptune, but larger on Uranus. We computed C/O latitudinal ranges and found C/O $\sim [0.06-0.52]$ on Uranus and $\sim [0.02-0.12]$ on Neptune. Comparison with a protoplanetary disk model suggests different formation or evolutionary pathways for the two planets. Our results highlight the dominant role of vertical mixing in constraining deep oxygen abundance and the importance of accounting for meridional variability and chemical uncertainties. This work also provides a framework for selecting the entry latitude of a future Uranus Orbiter and Probe mission.
Read the full abstract

All authors T. Briand, V. Hue, O. Mousis, T. Cavalié, T. Benest Couzinou, A. Schneeberger, M. D. Hofstadter

Robotics

FIRE-LIVWO: Robust LiDAR-Inertial-Visual-Wheel Odometry via Failure-Immune mmWave Radar Enhancement

Achieving robust SLAM in large-scale underground coal mines with complex structures and severe degeneracies remains highly challenging. Dense smoke and dust cause substantial loss of visual information and degrade LiDAR point-cloud features, while long, self-similar corridors induce geometric degeneration, leading to pronounced odometry drift. To address these issues, we propose FIRE-LIVWO: Failure-Immune mmWave Radar-Enhanced LiDAR-Inertial-Visual-Wheel Odometry, a tightly coupled multi-modal odometry framework based on an iterated error-state Kalman filter (IESKF). The framework fuses 4D mmWave radar, LiDAR, and visual features within a unified VoxelMap and jointly constructs LiDAR-radar point-to-plane residuals and sparse visual photometric residuals. In smoke-filled environments, we exploit the strong penetration of 4D mmWave radar and introduce pointwise Doppler velocity constraints to preserve state observability. In geometrically degenerate corridors, we tightly couple wheel odometry using non-holonomic constraints (NHC) and online lever-arm compensation to reduce drift. Our central contribution is a degeneration detection and adaptive fusion model switching strategy grounded in geometric and visual observability analysis, which quantifies observability online and dynamically adjusts modality weights. Real-world experiments in underground coal mines demonstrate that FIRE-LIVWO accurately identifies failure boundaries, enabling reliable modality switching under extreme conditions. Compared with baselines, it achieves superior accuracy and robustness (average localization error of 5.677m). We open source our code on Github to benefit the robotics community.
Read the full abstract

All authors Kun Hu, Menggang Li, Kaidi Wu, Zhiwen Jin, Yingjie Zhao, Chaoquan Tang, Eryi Hu, Gongbo Zhou

Robotics

RoboSPA: Can VLA Models Go Beyond Simple Scenes and Short-Horizon Tasks?

Vision-Language-Action (VLA) models have shown promising progress in language-conditioned robotic manipulation. However, existing datasets and benchmarks mainly evaluate task completion under predefined settings, offering limited insight into model reasoning under increasing spatial and procedural complexity. We introduce RoboSPA (Robot Spatial-Procedural Assessment), a large-scale robotic manipulation dataset and benchmark for diagnosing embodied reasoning in VLA models. RoboSPA focuses on two core dimensions, Fine-Grained Spatial Reasoning and Long-Horizon Procedural Planning, covering 10 task categories and 56 base tasks. Each task is instantiated across five difficulty levels, yielding 280 variants with increasing spatial ambiguity and procedural complexity. We collect 527K trajectories across multiple embodiments and diverse scenes. Beyond binary success rate, RoboSPA introduces diagnostic metrics for more detailed evaluation. Experiments on representative VLA models show that current systems still struggle with complex spatial relations, precise low-level execution, and memory-intensive planning. These results establish RoboSPA as a challenging diagnostic benchmark for developing more capable, reliable, and generalizable embodied agents. Our data and code are available at https://github.com/fanzhenxuan/RoboSPA.
Read the full abstract

All authors Zhenxuan Fan, Bo Zhang, Yutong Lin, Yuqian Yuan, Juekai Lin, Liang Liang, Zhuoyi Huang, Wenqiao Zhang, Juncheng Li, Siliang Tang, Jun Xiao, Yueting Zhuang

Computer Vision

Scalable Detection of Fossil Palynomorphs in Multifocal Digital Microscopy Images

Palynomorphs (microscopic, organic-walled fossils such as pollen, spores, and dinoflagellates) are important high-resolution records of past climates and are critical to the study of ancient ecosystems. Existing methods rely on manual analysis of high-resolution, multifocal digital microscopy images, which is slow and time-consuming and requires researchers to compromise on the scale of their investigations. To the best of our knowledge, our work proposes the first ever scalable end-to-end pipeline for automated palynomorph detection in whole slide images that addresses this bottleneck through: (1) efficient methods for decomposing and compressing digitized multifocal microscope slide images into tractable 2-dimensional tiles for analysis; (2) benchmarking modern object detection models, including RF-DETR, for the detection of palynomorphs, achieving an AP@50 of 0.879; (3) an efficient algorithm for the synthesis of detection outputs across large-scale, high-resolution images; and (4) an I/O optimization resulting in faster inference time. Our methods drastically reduce the time required for palynomorph detection in a single slide from often days of manual inspection to under one hour of automated analysis, enabling palynological research at a substantially greater scale.
Read the full abstract

All authors Abbas Shaikh, Praise Mayor, Patrick Ainlay-Vazquez, Aditya Viswanathan, Teon Golden, Eric Zhang, Ingrid C. Romero, Alexander E. White, Scott Wing, Arko Barman

Quantum Physics

Restricting the effects hides a nonphysical symmetry from every causal structure

Real-amplitude quantum theory is the subtheory of quantum theory invariant under complex conjugation, and experiments in networks of independent sources have measured correlations above the real bound. We construct a theory on which the same conjugation fails complete positivity and still leaves every probability in the unit interval, and whose correlations are exactly those of its own conjugation-invariant subtheory in every causal structure, the bilocality scenario included. Its states are all the density matrices, its effects are the operators every partial transpose of which is again a quantum effect, and its operations are the completely PPT-preserving maps. The symmetrized subtheory simulates it once each source carries a reference frame rather than each system. One map therefore receives three different verdicts in three theories, so the symmetry alone marks no boundary at all. Quantum theory admits every effect its states permit, and no theory with that property can hide a symmetry this way, so what sustains the separation measured in the network experiments is the absence of a restriction rather than any feature of conjugation. What does have a boundary is the class of theories whose correlations coincide with those of their symmetrized subtheory. We show that sectorial closure, meaning invariance of the effects and the operations under the symmetry acting independently on each source, suffices for the absence of a gap under any finite group, and that it cannot be weakened on the effects. Fixing unrestricted states and conjugation makes the effects of this theory the largest the symmetry admits and its operations the largest sectorially closed ones, so it is not one construction among several. Locating where sectorial closure fails, for a candidate effect set built from a fixed bound entangled state, is a finite computation on a single ray of effects.
Read the full abstract

Cosmology and Nongalactic Astrophysics

Supernovae Unite: Host-Galaxy Mass Measurements of Type Ia Supernovae and Their Impact on Cosmology

Current consensus suggests that Type Ia supernova (SN Ia) brightnesses post light-curve standardization correlate with their host-galaxy stellar masses, which must be accounted for to obtain accurate cosmological constraints. For example, Vincenzi et al. (2025) showed that different host-galaxy stellar mass measurements for the same dataset produce redshift-dependent differences of order $\sim 0.01$ mag, large enough to appreciably shift cosmological constraints. We present internally consistent host-galaxy stellar masses remeasured using aperture photometry and spectral energy distribution (SED) fitting for SN-Unite, which combines the spectroscopic Pantheon+ and the photometric Dark Energy Survey five-year (DES-SN5YR) samples into the largest SN Ia cosmology sample to date, with 2884 likely SNe Ia. We find that photometry and SED fitting choices shift SN-Unite Flat$w$CDM parameters well below statistical uncertainties. Our stellar masses differ from the Pantheon+ data release partly due to a redshift-dependent internal inconsistency within Pantheon+, while remaining largely consistent with the DES-SN5YR (DES-Dovekie) data release. When the Pantheon+ subsample of SN-Unite is combined with Baryon Acoustic Oscillations (BAO) and Cosmic Microwave Background (CMB) measurements, the significance for time-evolving dark energy increases from 3.4$\sigma$ to 4.0$\sigma$ based on the maximum a posteriori when our newly derived host-galaxy stellar masses replace the Pantheon+ data-release host-galaxy stellar masses, while DES-Dovekie remains virtually unchanged, consistent with the findings of Hoyt et al. (2026). By remeasuring the host-galaxy stellar masses using a consistent framework throughout the whole sample, we improve the robustness of the SN-Unite cosmological constraints against systematic differences in host-galaxy stellar mass measurements.
Read the full abstract

All authors J. Lee, R. Camilleri, T. M. Davis, D. Rubin, K. Bechtol, L. Galbany, M. Sako, M. Sullivan, T. E. Müller-Bravo, D. Scolnic, M. Vincenzi, D. Brout, C. Lidman, A. Möller, P. Shah, M. Acevedo, P. Armstrong, B. A. Bassett, R. C. Chen, H. T. Diehl, J. Frieman, K. Grech, B. Popovic, B. O. Sánchez, B. E. Tucker

2609.05321 · 4 Sep · 25 pages, 10 figures. Companion paper to Supernovae Unite: Combining Pantheon+ and DES-SN5YR. The host-galaxy stellar mass measurements and photometry will be publicly available upon acceptance of the paper. V2: For our Youtube Cosmology Talks, see https://www.youtube.com/watch?v=KT-j65gRnOQ. Removed inactive datalinks and fixed MAP significance. Submitted to the Astrophysical Journal

Computer Vision

Adaptive Gated Deepfake Detection for Low-Resolution and Resource-Constrained Environments

Deepfake detection models often rely on high-quality inputs, fixed inference paths, and computationally expensive architectures, limiting their use in low-resolution and resource-constrained settings. This paper proposes AdaGate-DF, an adaptive gated deepfake detection framework that uses image-quality cues to route samples through a dual multi-exit system so high-quality images can exit earlier and save compute. We evaluated AdaGate-DF against MaD-CoRN, DefakeHop++, and ShuffleNetV2 on two benchmark datasets (Celeb-DF and FaceForensics++) under multiple configurations to test image resolution dependence and training and inference efficiency. On Celeb-DF, AdaGate-DF achieves an AUC of 0.9370, outperforming MaD-CoRN and DefakeHop++ while maintaining a low inference latency. Resolution-based testing shows consistent improvement as input resolution increases, reaching an AUC of 0.9708 at 384 by 384. The FaceForensics++ results highlight that AdaGate-DF remains effective under class imbalance, following competitive results with evaluated models. Overall, AdaGate-DF demonstrated a practical balance between detection performance, uncertainty-aware prediction, and computational efficiency for variable-quality deepfake detection.
Read the full abstract

Strongly Correlated Electrons

Breakdown of Anomalous Hall Scaling in Dilute Kondo System

The confluence of strong electronic correlations and Berry curvature-driven transport constitutes a largely underexplored frontier in quantum materials research, particularly in systems where electronic correlations arise from flat-band physics rather than conventional $f$-electron states. Here, we report an intrinsic Berry curvature driven anomalous Hall transport in the dilute Kondo system Ru$_2$Mn$_{0.5}$Ti$_{0.5}$Ge, a non-$f$-electron Heusler alloy hosting flat bands with van Hove singularity proximate to the Fermi level. The anomalous Hall response is strikingly non-monotonic, violating the Fermi-liquid scaling relation near the magnetic transition and is restored only at low temperature, coincident with the onset of Kondo coherence. Ab initio calculations strongly establish that the Berry curvature originates from the Kondo hybridization induced interplay of flat bands and spin-orbit coupling mediated anticrossings at the Fermi level. Our findings demonstrate that Kondo coherence of flat bands bears a profound impact on Berry curvature associated ferroic responses, mandating a rigorous theoretical understanding of anomalous transport in the regime where reciprocal space topology and strong correlations are intrinsically intertwined.
Read the full abstract

All authors Arnab Bhattacharya, Prashant Singh, Ajay Kumar, Afsar Ahmed, Yaroslav Mudryk, I. Das, Anis Biswas

Machine Learning

Optimal Rates for Agentic Networked Information Aggregation

Building on the pioneering paper of Kearns, Roth, and Ryu (SODA'26), we study information aggregation in a networked learning model. The model captures a central pattern in agentic AI: each agent sees only part of the data and passes on only its own conclusion. Their model considers a linear regression problem with the mean squared error (MSE) loss. Agents sit in a DAG and each sees only a subset of the features and its parents' predictions, fits a linear predictor, and passes only its prediction forward. The benchmark is the full-feature learner that sees all raw features. A path of depth $D$ is $M$-covered if every block of $M$ consecutive agents collectively sees all raw features. Kearns, Roth, and Ryu proved that the excess mean squared error of the last agent on such a path is $O(M/\sqrt D)$, and gave a cyclic instance with excess error $\Omega(M/D)$ for $D<M^2$. We close this gap: the correct rate is constant up to depth $M^2$, and $\Theta(M^2/D)$ beyond it. We first give a sharper analysis of the cyclic instance and improve its lower bound to $\Omega(\sqrt{M/D})$ for $D<M^2$. We then construct, for every depth $D\ge M^2$, an $M$-covered path of depth $D$ with excess error $\Omega(M^2/D)$. The same instance gives the constant lower bound for all $D < M^2$. We also show that for any fixed distribution the excess error contracts geometrically along the path, ruling out any single instance that witnesses any polynomial lower bound at every depth. Finally, we prove the same optimal rate for logistic classification in the logit-passing model of Bateni et al., which considers the binary cross-entropy (BCE) loss. The same improved upper bound of $O(M^2/D)$ holds, and we transfer all the regression lower bounds by showing that on those examples the logistic path follows the least-squares path up to rescaling.
Read the full abstract

All authors MohammadHossein Bateni, Zahra Hadizadeh, MohammadTaghi Hajiaghayi, Mahdi JafariRaviz, Shayan Taherijam

Probability

Replica symmetry breaking at arbitrary depth

We study finite-step replica symmetry breaking (RSB) in mixed $p$-spin Ising spin glasses at positive temperature and zero external field. For every integer $k\geq1$, we construct polynomial mixtures whose Parisi measures are exactly $k$-RSB, with each such phase persisting on an open set of coefficients. More generally, for a broad class of polynomial covariances with an $r$-RSB Parisi measure, adding the interaction term $\lambda q^p$ for sufficiently large even $p$ produces a unique $r$-RSB-to-$(r+1)$-RSB transition as $\lambda$ varies near a critical value. Since such covariances exist for every $r\geq0$, this yields transitions of arbitrary depth. The main idea of the proof is that, for large $p$, the perturbation $\lambda q^p$ is negligible for overlaps bounded away from $1$, while its effect is concentrated at overlaps very close to $1$.
Read the full abstract

Quantum Physics

Compiling the 2D Fermi-Hubbard ground-state energy estimation algorithm for active volume quantum architectures

As quantum computing enters the early fault-tolerant era, circuit compilation choices will increasingly depend on details of the underlying architecture rather than solely optimizing for generic proxies such as non-Clifford count. We present an active-volume-aware compilation of the ground-state energy estimation algorithm for the two-dimensional Fermi-Hubbard model using quantum phase estimation and Trotterized time evolution. The proposed compilation reduces the active volume across $L\times L$ square lattices with $L=4$ to $20$, achieving up to a $3.9\times$ reduction over prior work optimized for non-Clifford cost. As a by-product of these compilation improvements, the resulting circuits also achieve state-of-the-art Toffoli counts, with a ~$2\times$ reduction for the $L=20$ case. Lastly, the active volume architecture and recent execution scheduling advances provide a means of translating these reduction trends into runtime. This demonstrates the increasing importance of architecture-aware compilation for practical early fault-tolerant quantum computing.
Read the full abstract

All authors Harriet Apel, Athena Caesura, Carys Harvey, Sam Heavey, Angus Kan, Jessica Lemieux, Ryan Levy, Sam Pallister, Joseph Peetz, William Pol, Sukin Sim, William A. Simon, Mark Steudtner, Gideon Uchehara

Nuclear Theory

Spectator shadowing as the source of the breaking of NCQ scaling at RHIC-FXT and FAIR energies

The apparent scaling of the elliptic flow of identified hadrons with the number of constituent quarks indicates the presence of quark coalescence, and in turn suggests the presence of a deconfined state in heavy-ion collisions. Recent measurements at RHIC found a breaking of NCQ scaling at low energies and suggested that this marks the onset of partonic collectivity at higher energies. Here, we present a new framework embedding the effect of spectator shadowing into the flow scaling relations. Using a toy model with an idealized quark coalescence source and a ballistic Glauber model for the bypassing spectator, we demonstrate that the observed breaking is an effect of spectator shadowing.
Read the full abstract

Artificial Intelligence

Large Language Models for HVAC Operations in Building Energy Systems: A Critical Review of Methods, Applications, and Deployment Readiness

Building automation systems generate rich sensor data yet remain insight-poor because heterogeneous point naming, missing metadata, and fragmented documentation obstruct their operational use. This systematic review analyses and codes 66 peer-reviewed studies on large language models (LLMs) for HVAC operations published between 2023 and March 2026. Each study is classified across five application families and three LLM method families and assessed for evidence realism, deployment readiness, and the responsibility boundary between the LLM and physical HVAC decisions. The corpus is concentrated in building energy modelling (BEM, 32 of 66 papers), while load forecasting remains too sparse for subfield-level conclusions. Only four studies reach pilot-level evidence, and none reports sustained operational deployment. No study was classified as ready-now for industry adoption; three were near-term and 63 research-only. Nevertheless, several bounded, human-in-the-loop uses merit near-term trials, including point-name normalisation, document-grounded operator support, BEM workflow assistance, and advisory interfaces around physics-based controllers. Conventional machine learning (ML), model predictive control (MPC), reinforcement learning (RL) and ontology-based tools remain more adopted for high-frequency control, short-horizon numerical forecasting, and well-posed ontology mapping, while autonomous agentic operation and unvalidated occupant proxies remain research-stage. Current evidence therefore supports LLMs primarily as semantic and workflow layers rather than autonomous HVAC controllers. Future work should prioritise field-validated benchmarks, orchestration evaluation under operational constraints, and LLM-MPC/RL architectures with bounded latency and verifiable safety properties.
Read the full abstract

All authors Alexander Neubauer, Tianzhen Hong, Han Li, Mengbo Yu, Amin Darbandi, Yannick Fürst, Martin Kriegel

Quantum Physics

SAR and InSAR Change Detection with Quantum Generative Models

Change detection in synthetic aperture radar (SAR) and interferometric synthetic aperture radar (InSAR) underpins disaster response, infrastructure monitoring and land-use enforcement. Detection is limited by the background estimator, which conventionally forms a conditional expectation directly from observed pixel statistics and degrades where those statistics are sparse, including the regime produced by the heavy-tailed marginals of sub-meter-resolution radars. In this work, we integrate state-of-the-art satellite imagery with quantum machine learning on IonQ trapped-ion-based quantum processors. By replacing the empirical conditional with a quantum circuit Born machine (QCBM)-sampled generative model in Copula space, we substantially improve change detection on sparse real-world images. On Capella Space satellite image acquisitions, the generative estimator matches conventional methods when the observed statistics are adequate, and substantially outperforms them when they are not. Executing the trained model on IonQ trapped-ion based hardware reproduces the results of the ideal and noisy simulations and demonstrates up to par, or even better, performance with the classical state-of-the-art methods. For a SAR dataset of an airport, QPU circuit evaluations for both training and inference achieved a maximized filtered F1 score of 0.32, compared with 0.16 and 0.24 for the two classical baselines. For an InSAR dataset of a volcanic lava flow, all three methods reached a maximum filtered F1 of approximately 0.66. These experiments demonstrate the feasibility of executing a QCBM-based background estimator on trapped-ion hardware. We further demonstrate that the QCBM method successfully extends to interferometric coherence data, achieving performance comparable to classical approaches.
Read the full abstract

All authors Samwel K. Sekwao, Shaunak De, Alexis Hocken, Scott Staniewicz, Evgeny Epifanovsky, Craig Stringham, Gordon Farquharson, Martin Roetteler, Panagiotis Kl. Barkoutsos, Jason Iaconis

Earth and Planetary Astrophysics

Evidence for LP 890-9d via Transit Timing Variations

LP 890-9, also known as SPECULOOS-2 and TOI-4306, is a nearby late-M dwarf hosting two confirmed transiting rocky exoplanets. We analyze 20 JWST/NIRSpec PRISM transits of LP 890-9b and LP 890-9c obtained as part of GO program 7073 and detect statistically significant transit timing variations (TTVs), with peak-to-peak amplitudes of ~17 s and ~35 s, respectively. Using analytic linear TTV theory, we find that the known two-planet configuration cannot reproduce the measured TTV amplitudes or super-period, whereas three-planet models provide substantially better fits. The best-fit configuration places the candidate third planet, LP 890-9d, between planets b and c, with an orbital period of ~4.4 days; however, the current data do not uniquely determine its orbital architecture, and periods spanning 4.0-6.9 days remain plausible. TESS is insensitive to transits of LP 890-9d and we find no evidence for the candidate in JWST observations, although the phase coverage (ranging from ~50% to ~80%) depends strongly on the candidate orbital period. Additional high-precision transit observations of LP 890-9b and LP 890-9c are needed to refine their TTV solutions and further constrain the orbital properties of the third planet.
Read the full abstract

All authors Kevin B. Stevenson, Guangwei Fu, Kristin S. Sotzen, E. M. May, Larissa Palethorpe, Jacob Lustig-Yaeger, Ted M. Johnson, Eric Agol, Katherine A. Bennett, Carlos Gascón, David K. Sing, Jeff A. Valenti, Hannah R. Wakeford

Number Theory

Distribution of points near the origin in the $d$-dimensional Lagrange spectrum

We develop a new framework, inspired by Schmidt's games, to study the Lagrange spectrum for simultaneous Diophantine approximation in dimension $d\geq 2$. We show the Hausdorff dimension of the set of points in $\mathbb{R}^{d}$ whose best approximation constant lies in $[\varepsilon,\varepsilon(1+\delta\varepsilon^{d})]$, for some constant $\delta>0$, is positive, and for a slightly larger set approaches full dimension as $\varepsilon \to 0$. The proof combines a novel application of the Simplex lemma near rational points with the game-theoretic framework. We also use an elementary observation to show that the naturally defined Lagrange spectrum for systems of linear forms is uncountable in the case of square matrices.
Read the full abstract

Game Theory

Closing Gaps in Online Fair Division

We study the online fair division of indivisible items, where items arrive one at a time and must be allocated immediately and irrevocably. We address three central open questions in the literature. First, we show that for every $n\ge 2$ agents, every fixed $k\ge 1$, and every $\alpha\in(0,1]$, no online algorithm can guarantee $\alpha$-PROP$k$ against an adaptive adversary. This remains true even when the total number of goods is known in advance, all values lie in $[0,1]$, and every good is positively valued by at most two agents. The impossibility extends to a broad range of standard envy-based, proportionality-based, and share-based fairness notions. We also establish an analogous impossibility for chores. Second, in the setting with predictions, a lightweight form of future information, the maximum item value, was previously known to guarantee only $1/n$-PROP1, leaving open whether the dependence on $n$ is necessary. We give a deterministic $19/30$-PROP1 algorithm against adaptive adversaries that does not require knowing the total number of goods. Given an additional upper bound $\kappa\in[2,n]$ on the number of agents who value any good positively, the guarantee improves to $\max \{ 19/30, n/(n+\kappa) \}$. The guarantee remains a positive constant under any fixed one-sided prediction error below one. When predictions are exact and the total number of goods $m\ge n\log n$ is known in advance, a deterministic algorithm achieves the same $19/30$-PROP1 factor together with $O(\sqrt{m\log n/n})$ maximum additive envy after normalizing each agent's values by their maximum item value. Third, against a non-adaptive adversary, we determine the tight high-probability PROP1 guarantee of the classical Like rule, which assigns each good uniformly among the agents who value it positively. Its guarantee improves when fewer agents value the same good, unlike uniform random allocation.
Read the full abstract

Machine Learning

How Does mHC Use Its Residual Streams? Selective Routing and Near-Identity Mixing

Hyper-Connections and their manifold-constrained variant mHC widen a residual pathway from one stream to n, yet how trained models use this capacity remains unclear: how broadly blocks read and write, how strongly the residual pathway mixes streams, and whether the streams carry distinct representations. We examine these properties in the four-stream residual pathway of DeepSeek-V4-Flash using effective stream counts, cross-stream residual weights, and inter-stream cosine similarity. Read/write routing is concentrated but varies across depth: a typical attention or FFN site effectively uses about two streams, while the dominant stream changes across layers and the representations remain directionally distinct. Residual mixing is modest and occurs primarily in early layers; in layers 22-42, the pathway mostly carries each stream forward separately. Targeted interventions establish the functional significance of these patterns. Replacing the late mixers by identity increases C4 perplexity by only 1.9% and preserves the six-task average score, whereas replacing the early mixers increases perplexity by 41%. Fixing each early mixer to its C4 diagnostic mean increases perplexity by only 0.2% and reduces the average score by 0.25 percentage points, showing that its site-specific structure matters more than its token-wise variation on the evaluated metrics. Likewise, retaining the three largest routing weights per token at every site increases perplexity by at most 2.7% and changes the average score by at most 0.4 points. Thus, the studied model realizes only part of the flexibility afforded by four-stream mHC: individual blocks rarely require all four streams, and late residual mixing provides little measured benefit.
Read the full abstract

All authors Pengxiang Zhao, Xing Li, Xianzhi Yu, Wei Guo, Zhenhua Dong

Statistics Theory

Subsampled Pseudo-posteriors for Scalable Bayesian Moment-condition Inference

Bayesian pseudo-posteriors based on moment conditions, such as Bayesian empirical likelihood (EL) and Bayesian exponentially tilted empirical likelihood (ETEL), provide a robust route to Bayesian inference when the model is specified only through moment restrictions, but their computation is often prohibitive. In this work, we address two barriers to Bayesian pseudo-inference under moment restrictions. First, scalable sampling is challenging for tall data because the pseudo-likelihoods are not log-additive across observations, so standard likelihood-subsampling methods do not apply. Second, in simulation-based and latent-variable problems, the moment function itself may be defined through an intractable expectation, making full-data pseudo-posterior inference infeasible. We develop a new family of subsampled pseudo-posteriors that replaces the full-data pseudo-posterior with an aggregate of mini-batch pseudo-posteriors. We show that naive mini-batching induces a shifted-mixture distortion, yielding aggregates that are correctly centered but overly diffuse. To remove this distortion, we introduce moment-function-level control variates that align mini-batch moment equations around a reference estimator and recover the full-data posterior shape. For Bayesian ETEL, we establish finite-sample total variation bounds to the full-data pseudo-posterior, allowing discontinuous and intractable moment functions. Numerical experiments show that the method applies broadly across moment-condition-based Bayesian pseudo-inference, closely approximates full-data pseudo-posteriors, and substantially reduces computational cost.
Read the full abstract

Systems and Control

Forecast-Ensemble-Based Active Binary-Threshold Query Design for Interval Data Assimilation

Data assimilation estimates the evolving state of a dynamical system by combining model forecasts with observations. While many conventional methods assume point-valued measurements, practical sensing systems may instead provide coarse information such as binary, ordinal, inequality, or interval-valued reports. Interval Data Assimilation (IDA) provides a principled framework for assimilating such information, but assumes that the observations to be assimilated are specified in advance. In many sensing settings, however, both where to query and what inequality to ask can be chosen, while only a limited number of queries can be issued. This raises a new observation-design problem: how should informative inequality queries be selected from the forecast uncertainty before their responses are known? We propose Active Query Design for Interval Data Assimilation (AQD-IDA), which uses the forecast ensemble to design binary threshold queries prior to the IDA update. AQD-IDA treats both the observation target and the threshold defining the inequality as design variables, thereby allowing the assimilation system to determine not only where to observe but also what question to ask. We develop query-selection criteria that account for forecast uncertainty, redundancy among queries, and anticipated reduction in posterior uncertainty. Experiments on Lorenz–96 and a spherical quasi-geostrophic model show that adaptive query design improves assimilation accuracy under a fixed binary-query budget, with joint observation-target–threshold selection outperforming fixed or separately designed queries. These results demonstrate that actively designing the information supplied to data assimilation can substantially increase the value of coarse observations.
Read the full abstract

High Energy Physics (Phenomenology)

Chaos is Target-Blind in High-Energy QCD Evolution

Chaotic evolution measures how rapidly nearby configurations separate, but the existence of such rapid evolution does not by itself determine what physical information the evolution actually carries. For fixed-coupling JIMWLK evolution with fresh transverse-local, color-isotropic noise, we show that this distinction becomes exact - leading to a no go theorem. At finite rapidity step and finite transverse regulator, the exact linearized update depends on the Wilson-line background only through local adjoint rotations of fresh color-isotropic noise. Once the past is conditioned upon, these rotations leave the complete local noise measure invariant, so the tangent process has the same probability law for any target ensemble evolved with the same regulated fixed-coupling dynamics, including dilute and saturated ensembles. After the exact global color-rotation mode is removed, the reduced tangent process found to have a positive leading Lyapunov exponent which indicate chaotic evolution. Fixed-coupling JIMWLK therefor can be dynamically chaotic while its tangent-only statistics carry no information about the target ensemble. Spatially nonlocal noise however evades the argument because correlations between distinct sites leave relative Wilson lines in the conditional covariance.
Read the full abstract

Number Theory

F-sets of arbitrary finite width

Ferraguti and Micheli conjectured that non-trivial $F$-sets of every prescribed width exist over each finite field. We prove the finite-width part of their conjecture over $\mathbb{F}_q$ for every $q\neq2,3$. Fix a suitable prime $\ell$. For a degree bound $D$, we consider the saturated family of all irreducible polynomials $g(X^{\ell^j})$ whose core $g$ has degree at most $D$. A factor-descent lemma shows that every irreducible factor arising from a shifted difference belongs to the same family and has strictly smaller core degree. The saturated family is therefore an $F$-set of finite width. Dirichlet's theorem for $\mathbb{F}_q[X]$ produces core ladders of arbitrary length, and Kummer lifting reproduces each ladder at infinitely many scales. These parallel ladders force the width to be sufficiently large; an appropriate tail of the nullity filtration then has the prescribed width.
Read the full abstract