Dynamical Systems
Survivor Set of Gauss Map with a Hole at $1$
Cheng Liu and Lingmin Liao
Let $T$ be the Gauss map. For each $\alpha\in [0,1]$, let $E(\alpha)$ be the survivor set consisting of all $x\in [0,1]$ whose orbit $\{T^k(x):k\ge 0\}$ never enters the interval $(\alpha,1]$. We prove that the set-valued function $\alpha\mapsto E(\alpha)$ is increasing and locally constant almost everywhere. Moreover, we explicitly characterize all intervals where the function $\alpha\mapsto E(\alpha)$ remains constant. We also describe the points at which the map $\alpha\mapsto E(\alpha)$ is not locally constant. Further, we obtain the Hausdorff dimension of $E(\alpha)$ for all $\alpha\in [0,1]$. We prove that the function $\alpha\mapsto \dim_H(E(\alpha))$ forms a devil's staircase on $(0,1]$, but is not continuous at $0$. This discontinuity implies that the result of Raith [Acta Math. Univ. Comenian. 63 (1994), 39-53] fails for piecewise monotone interval maps with infinitely many branches. Finally, we give an application quantifying the set of points sharing identical $\alpha$-continued fraction expansion and regular continued fraction expansion.
Read the full abstract ▾
Quantum Physics
Sample-Size Scalability and Representation Sensitivity of Quantum Machine Learning Models for Higgs Boson Event Classification
Shiksha Bhardwaj and Ankhi Roy
Quantum machine learning offers new approaches to multivariate event classification in high-energy physics, but its performance relative to optimized classical methods remains uncertain under limited training data and restricted
quantum resources. We investigate the sample-size scalability and representation sensitivity of a
quantum support vector machine and a variational
quantum classifier for
Higgs-
boson signal–background discrimination using simulated events from the ATLAS
Higgs Machine Learning Challenge. The models are trained on balanced subsets ranging from 250 to 4000 events and evaluated on held-out samples preserving the original class distribution. Principal component analysis (PCA) and autoencoder (AE) representations are examined to reduce the input dimensionality and
quantum-resource requirements. Four optimized classical classifiers are used as
benchmarks under a common preprocessing, validation, and evaluation protocol. We assess training-sample scalability in classical and
quantum classifiers using a four-component PCA representation and examine representation sensitivity through a secondary matched-event comparison of four- and eight-component PCA representations. Within the evaluated
Higgs-
boson classification setting, the QSVM showed stronger performance than the VQC at larger training sample sizes, while remaining below the best-performing classical classifiers.
Read the full abstract ▾
Operating Systems
GMM: Generalized OS Memory Management for Programmable XPUs
Weixi Zhu, Bin Wang, Zhiyuan Fang et al. (+5 more)
Since the AI boom, the fast evolving programmable XPU devices utilize high-bandwidth local
memory to accelerate domain-specific tasks offloaded from the host CPU, resulting in a proliferation of XPU
memory management systems implemented in proprietary drivers. However, existing OS core
memory management (MM) continues to lack a high-level software abstraction, excluding the XPUs from leveraging OS advanced
memory features. The XPU manufacturers have to reimplement independent XPU MM drivers, exacerbating the fragmented XPU software ecosystem. This paper presents GMM, generalized
memory management that provides unified MM subsystems for XPUs. By adding an explicit logical mapping abstraction, GMM 1) provides universal NUMA-aware
memory API for XPUs; 2) allows ease of XPU MM delegation; 3) enables sharing of advanced XPU
memory features. Lightweight XPU MM drivers are implemented for two representative XPU manufacturers, Nvidia and Huawei. Both GMM-based drivers obtain competitive performance compared with their vendor-supplied drivers. GMM also obtains up to 41% higher fault-driven page migration throughput. Two real-world industrial case studies are exercised, where GMM reduces XPU
memory utilization by 10% with up to 22.6x faster inference serving switch latency.
Read the full abstract ▾
All authors Weixi Zhu, Bin Wang, Zhiyuan Fang, Jiahua Wang, Yao Guo, Gingfung Yeung, Adam Barker, Xinwei Hu
Computation and Language
LongHisDoc: A Comprehensive Benchmark for Chinese Long Historical Document Understanding
Zeyu Shan, Yongxin Shi, Shihang Wu et al. (+4 more)
Chinese historical documents are invaluable cultural treasures, but their complexity poses significant comprehension challenges for non-specialists. Despite the success of Large
Language Models (LLMs) and Large
Vision-
Language Models (LVLMs) in modern document tasks, their capabilities in Long (full-book) Historical Document understanding remain largely unexplored. To bridge this gap, we introduce LongHisDoc, the first
benchmark tailored for long historical document understanding. LongHisDoc comprises 1,012 expert-annotated QA pairs derived from 101 PDF documents, with an average of 85.1 pages and 42,776.4
tokens per document. It spans 10 categories and features four QA types: single-page, cross-page, multi-hop, and unanswerable. Answering these questions requires integrating
multimodal evidence from text, layout, and visual elements. Our evaluation reveals that the top-performing LVLM and LLM achieve only 62.9% and 69.9% accuracy, respectively, highlighting a substantial performance gap. Furthermore, we assess various modal retrievers and the end-to-end RAG performance to provide references for
Retrieval-Augmented Generation (RAG) systems. We believe that LongHisDoc serves as a comprehensive
benchmark to catalyze future research in full-book historical document understanding. The
benchmark and evaluation code are available at https://huggingface.co/
datasets/qweq12433454/LongHisDoc and https://github.com/SCUT-DLVCLab/LongHisDoc.
Read the full abstract ▾
All authors Zeyu Shan, Yongxin Shi, Shihang Wu, Peirong Zhang, Yuanyuan Fang, Zhaopeng Wang, Lianwen Jin
Robotics
Who to Trust in Multi-Robot Systems: Joint Robot and Object Hypothesis Trust Inference via Neural-Symbolic Cross-Validation
Ruiyang Wang, Hao-Lun Hsu, Haocheng Meng et al. (+2 more)
Multi-
robot systems (MRS) offer significant advantages over single-
robot platforms by enabling information sharing and coordination for different tasks in complex environments. However, reliable cooperation critically depends on inferring trust over both information sources and their reported hypotheses, particularly under partial observability and potential
adversarial interference. Existing approaches frequently depend on pre-established trust relationships or side-channel information, which may not be available or reliable in realistic MRS deployments. This work introduces NeST-Bayes, a neural-symbolic framework that jointly infers
robot and hypothesis trust using only inter-
robot observation consistency, without requiring preassigned
robot trust, reputation information, or side-channel identity signals. NeST-Bayes infers instantaneous trust evidence scores for
robots and object hypotheses by
reasoning over ego-centric relational graphs constructed from cross-
robot observation consistency. These evidence scores are accumulated over time through
Bayesian updates of Beta-distributed trust beliefs to maintain temporally consistent trust estimates. Extensive evaluations show that NeST-Bayes achieves the highest mean object precision and
robot recall across all
benchmark settings, and the highest
robot precision and object recall in most settings, compared to trust-based methods. Under aggressive optimized attacks, it achieves 98.8%
robot precision with 93.8% recall and 98.3% object precision with 95.4% recall. In an occlusion-rich Webots warehouse, it maintains 92.5%
robot precision with 92.0% recall and 97.3% object precision with 85.0% recall. These results demonstrate joint improvements in the reliability and coverage of retained information.
Read the full abstract ▾
All authors Ruiyang Wang, Hao-Lun Hsu, Haocheng Meng, Songqiao Xie, Miroslav Pajic
Analysis of PDEs
Domain dependence of Sobolev-curl constants and structure of critical curl-curl solutions
Yong Liu, Jun Wang, Xuan Wang et al. (+1 more)
This paper investigates intrinsic critical Sobolev-curl constants on bounded domains and the structure of finite-energy solutions to the critical curl-curl equation in R^3. First, for 1< p < 3, we prove that the intrinsic critical p-curl constant on (0, a)\times D is bounded above by a^{p/3} times a Sobolev quotient on D. The proof uses an aligned test field whose optimal curl-free correction vanishes. Consequently, these constants tend to zero along thin convex boxes. For p = 2, we construct explicit convex and nonconvex domains with intrinsic constant below the sharp scalar Sobolev constant. The nonconvex example gives an affirmative answer to the domain-existence question in Problem 5 of Mederski and Szulkin(Arch. Ration. Mech. Anal. 241 (2021), 1815-1842), while the convex example disproves the lower bound stated there for convex domains. Concerning their Problem 2, we obtain an implicit Newtonian representation of finite-energy solutions and exclude two special forms of solutions. These structural results do not yet provide an explicit ground state or an unconditional ground-state decay rate as asked in Problem 2.
Read the full abstract ▾
All authors Yong Liu, Jun Wang, Xuan Wang, Wen Yang
Signal Processing
Performance Evaluation of Mobile Target Sensing and Tracking Across Multiple Sensing Modes
Necati Kagan Erkek, Marco Di Renzo, Arman Shojaeifard et al. (+3 more)
Moving extended-target sensing and tracking are studied in a beamformed PRS/OFDM
ISAC framework. The target is modeled as a rigid cluster of scatterers whose Tx–target–Rx echoes coherently combine in a
MIMO scattering channel. Delay, azimuth/elevation, and Doppler are estimated using PRS matched filtering with delay centroiding, single-cluster 2D-MUSIC, and FFT-based Doppler estimation, then used for independent localization and EKF-based tracking. The framework is evaluated in monostatic and bistatic modes under target-directed, receiver-directed, and omnidirectional beamforming. Simulation results show that EKF tracking reduces the aggregate 3D RMSE from 1.61 m to 1.38 m relative to independent localization. Target-directed beamforming substantially improves tracking accuracy, achieving 1.38 m RMSE compared with 5.91 m and 8.38 m for omnidirectional and receiver-directed beamforming, respectively. The results highlight the importance of target-aware beam management, delay-angle-Doppler fusion, and sensing geometry for mobile-target tracking in
ISAC networks.
Read the full abstract ▾
All authors Necati Kagan Erkek, Marco Di Renzo, Arman Shojaeifard, Kunjan Shah, Mohammad Heggo, Yasser Mestrah
Statistical Methodology
Identification and Estimation of Treatment Effects from Network Data with Confounded Social Links
Licheng Liu
In observational network data, the adoption of a treatment and the formation of social ties often depend on the same characteristics that affect outcomes. We propose a generalized propensity score for each unit's effective treatment, built from a model for treatment assignment and a dyadic model for network formation, which may include latent node heterogeneity. The score is marginalized over the network given
covariates rather than conditioned on the realized network, and the estimand is a population effect: the average effect of an exposure level for a unit drawn from the population of units and networks. We relate this estimand to those of methods that condition on the realized network and show what marginalizing costs and buys. It requires network formation to be unconfounded given observed
covariates, a stronger assumption than unconfoundedness of treatment given the network. In exchange, it defines effects at exposure levels that poorly connected units cannot reach, and at levels every unit can reach it gives an
estimator whose asymptotic variance is no larger than the network-conditional one. We combine the score with an augmented inverse probability weighting
estimator, prove asymptotic normality as the number of networks grows with bounded network size, discuss what carries over to a single large sparse network, and apply the method to the effect of a political communication technology on political participation in Uganda.
Read the full abstract ▾
High Energy Astrophysical Phenomena
Impact of Decay-Chain Feeding by Long-Lived Heavy Nuclei on Nuclear Cosmochronology
Qiu-Hong Chen, Meng-Hua Chen, Xianlin Yang et al. (+1 more)
Nuclear cosmochronology provides an independent method for determining the ages of astrophysical objects without relying on
Galactic chemical
evolution models. Long-lived radioactive actinides synthesized via the rapid
neutron-capture process ($r$-process) particularly thorium and uranium serve as ideal chronometric
nuclei. However, predicting the initial abundances of these
nuclei and their subsequent long-term
evolution remains a major source of uncertainty in
nuclear cosmochronology. In this work, we investigate the impact of decay-chain feeding from long-lived progenitor
nuclei on
nuclear cosmochronology and estimate the corresponding age corrections. We develop a self-consistent radioactive decay network that includes all relevant decay pathways and couple it to $r$-process nucleosynthesis calculations. By tracking the long-term
evolution of $r$-process nucleosynthesis based on
neutron star merger trajectories, the dominant long-lived
nuclei shaping the
evolution of the chronometric elements Th and U are identified as $^{232}$Th, $^{235}$U, $^{236}$U, $^{238}$U, and $^{244}$Pu. It is found that delayed feeding substantially modifies the long-term
evolution of actinides particularly for $^{232}$Th, which receives significant late-time contributions from heavier progenitor
nuclei such as $^{244}$Pu and $^{236}$U. As a consequence, the U/Th and Th/X chronometers exhibit significant sensitivity to delayed feeding from the progenitor
nuclei $^{244}$Pu and $^{236}$U, with age corrections reaching $\sim5$~Gyr. These results suggest that decay-chain feeding from long-lived progenitor
nuclei plays a crucial role in
nuclear cosmochronology and should be properly accounted for in age determinations based on actinide chronometers.
Read the full abstract ▾
All authors Qiu-Hong Chen, Meng-Hua Chen, Xianlin Yang, En-Wei Liang
Artificial Intelligence
Energy Ring: Stabilizing Feature Energy for Reliable Test-Time Discovery
Tianle Liu, Fan Lyu, Tingliang Feng et al. (+6 more)
Test-Time Discovery (TTD) aims to recognize known classes and detect novel ones while the model is already deployed, operating directly on a continual unlabeled test-time stream. Existing TTD methods are overly conservative, making known and unknown features collapse together and difficult to separate during test-time adaptation. In this paper, we revisit TTD from an energy perspective and find that the natural energy gap between known and unknown classes is unstable and easily collapses. We therefore propose Energy Ring to stabilize this gap by anchoring the two types of classes on separate energy rings. A lightweight adapter enlarges the norm-based gap while preserving known-class directions, and a perturbation consistency module forms compact clusters for unknown classes. During testing, energy-guided confidence weighting improves prediction, and a stability-based rectification mechanism corrects unreliable updates. Experimental results show that Energy Ring significantly improves TTD performance against existing TTD methods.
Read the full abstract ▾
All authors Tianle Liu, Fan Lyu, Tingliang Feng, Chenggong Ni, Tianqi Wang, Zhang Zhang, Fuyuan Hu, Rui Yao, Tao Zhou
Optimization and Control
Consumption adjustment with mean-reverting risk premia: A singular stochastic control approach
Gugyum Ha, Junkee Jeon and Jehan Oh
We study a two-sided singular
stochastic control problem in which changes in consumption carry utility costs and the risk premium follows the mean-reverting Kim-Omberg model. Optimal consumption is reflected between two moving barriers given by thresholds that depend on the current risk premium. These thresholds are the stopping boundaries of a Dynkin game for a two-dimensional
diffusion driven by a single Brownian motion. A sharp occupation estimate shows that they are Lipschitz in logarithmic scale, uniformly in the cost of reducing consumption, and that smooth fit holds along the
diffusion direction at every contact point. Outside a bounded interval of risk premia and at most two further points, the thresholds are infinitely differentiable. As the reduction cost grows, optimal consumption converges to the ratcheting policy.
Read the full abstract ▾
Artificial Intelligence
ScienceBuddy-PV: A Verifier-Governed Recursive Improvement Protocol for Panvascular Intervention Research
Lingsen You, Yujun Guo, Xinyu Zhong et al. (+4 more)
ScienceBuddy-PV adapts the ScienceBuddy recursive-in-recursive design to panvascular intervention research through a structured evidence ledger, device-vessel ontology, abstention contract and retained failure bank. Recursive harness optimization and outer-loop model learning, including reinforcement learning, are specified but not executed. We audit a fixed component, rather than the complete
agent system, using separate ARCADE syntax and stenosis tasks (1,000 training, 200 validation and 300 test images per task). A shared task-masked convolutional model was trained once per representation setting; its parameters and the heuristic support gate were then held fixed, with the stenosis threshold selected on validation data. On the locked test sets, syntax Dice was 0.475 (95% CI 0.459-0.490) and stenosis Dice was 0.231 (95% CI 0.214-0.249). Stenosis pixel AUROC was 0.957 and AUPRC was 0.180 at prevalence 0.00975. The gate accepted 76.7% of stenosis-frame outputs versus 100% for a score-only comparator. In a constructed mask audit, it accepted 44.3% of reference candidates and rejected 55.7%, exposing the cost of abstention. Acceptance does not establish correctness or clinical safety. This public-data component audit makes support and abstention measurable and provides a reproducible protocol for future collaborative
agent studies. It does not demonstrate recursive improvement, patient-level validity or clinical device recommendations.
Read the full abstract ▾
All authors Lingsen You, Yujun Guo, Xinyu Zhong, Zijie Feng, Wentong Wang, Li Shen, Junbo Ge
Robotics
Safety-Critical Quadrotor Navigation Using Fourth-Order Control Barrier Functions with Full Nonlinear Quadrotor Dynamics
Yanze Liu, Yaşar İdikut, Dhruv Madhusudan Agrawal et al. (+2 more)
Safety-critical quadrotor
navigation in cluttered environments typically relies on hierarchical architectures in which a high-level safety filter modifies position references that are then tracked by a separate attitude controller. However, this separation can create a mismatch between translational safety objectives and attitude
evolution. To address this gap, we propose a safety controller based on fourth-order control barrier functions (CBFs) that explicitly incorporates the full nonlinear quadrotor dynamics, including attitude dynamics and torque-level actuation. Furthermore, we apply the integral high-order CBF technique to resolve the mixed-relative degree issue in the derived constraint formulation. The resulting formulation enforces obstacle-avoidance constraints while optimizing control torques, granting the safety layer direct authority over attitude behavior and eliminating the hierarchical disconnect inherent in reference-tracking safety filters. We validate the approaches in both high-fidelity simulation and real-world flight experiments, demonstrating reliable constraint satisfaction and successful obstacle avoidance. These results suggest that torque-level high-order CBFs provide a practical pathway to provably safer quadrotor
navigation under full rigid-body dynamics.
Read the full abstract ▾
All authors Yanze Liu, Yaşar İdikut, Dhruv Madhusudan Agrawal, Wei Xiao, Guanrui Li
Chemical Physics
Learning Electron Density with Virtual Atoms for Transferable Atomistic Representations
Yuzhi Liu, Xinzijian Liu, Chengqian Zhang et al. (+3 more)
Electron density is a fundamental quantity of electronic structure, yet most machine learning density models treat it solely as a reconstruction target and are developed separately from machine learning interatomic potentials. Here, we introduce a virtual-atom formulation within the DPA3 framework, in which density query points are represented as virtual nodes and processed by the same parameterized encoder used for real atoms. An asymmetric neighbor construction allows virtual atoms to receive information from nearby real atoms without altering the real-atom interaction graph. The resulting framework supports density, energy, and force prediction through task-specific readout heads. The proposed model achieves state-of-the-art density reconstruction accuracy on the Materials Project
benchmark, matches the best reported performance on nickel-manganese-cobalt battery cathode materials, and is competitive with dedicated density models on QM9. When density labels are used jointly with energy and force supervision, the model improves in-domain accuracy, makes local environment information more accessible under linear probing, and reduces zero-shot force errors on larger QM40 molecules. Density-only checkpoints also provide useful initializations for downstream energy-force learning under the fixed fine-tuning protocol. Together, these results support electron density as a source of electronic-structure supervision for interatomic-potential representations.
Read the full abstract ▾
All authors Yuzhi Liu, Xinzijian Liu, Chengqian Zhang, Anyang Peng, Duo Zhang, Han Wang
Machine Learning (Statistics)
Identication and Estimation of Nonsymmetric Determinantal Point Processes
Christian Gouriéroux and Yang Lu
Determinantal point processes (DPPs) provide tractable distributions for nite random sets, but the usual symmetric-kernel specication imposes nonpositive pairwise dependence. Statistical research on nonsymmetric DPPs, which can accommodate positive dependence, remains limited. We study identication and estimation for this class. First, we establish the equivalence of two sucient conditions for a nonsymmetric kernel to dene a valid DPP. Second, we develop the identication theory for nonsymmetric DPPs. Third, we develop cardinality-truncated maximum-likelihood estimation and estimation by averaged
stochastic gradient descent. Under suitable identication and regularity conditions, we establish consistency and asymptotic normality for both
estimators. Simulations examine identication, the eects of truncation, and the nite-sample performance of the
estimators. We illustrate the methods with two cyber applications at markedly dierent sampling frequencies: weekly sector-by-breach data and second-by-second enterprise authentication data with more than ve million set observations.
Read the full abstract ▾
Machine Learning
Query-Adaptive Neuron Steering for LLMs
Yanxi Pan, Jingwen Wu, Yan Ma et al. (+8 more)
Effectively eliciting the capabilities of large
language models (LLMs) at inference time remains an important challenge. Existing studies address this challenge through context engineering, which shapes user queries, or fixed steering, which directly intervenes in model computation. However, predefined contexts and fixed intervention strategies may not adequately accommodate the diverse knowledge and
reasoning requirements of individual queries. This raises a natural question: Can internal steering in LLMs be made query-adaptive? In this work, we investigate whether query-induced internal activations can guide adaptive steering in LLMs. Our pilot studies reveal subject- and layer-dependent
neuron activation patterns and show that amplifying subject-relevant
neurons can improve model performance. Based on these findings, we propose Q-NeuS, a query-adaptive
neuron steering framework that dynamically determines which
neurons to modulate and by how much for each input. Specifically, Q-NeuS constructs a compact subject-guided candidate
neuron space and employs a lightweight Query-Adaptive Gate to generate
neuron-wise modulation coefficients. To learn effective intervention strategies, we introduce Neu-GRPO, which adapts group-relative optimization to train the Query-Adaptive Gate using the relative performance of multiple
neuron intervention configurations. The base LLM remains frozen throughout training and inference, and no task or subject labels are required at inference time. Extensive experiments across multiple LLMs and
benchmarks demonstrate that Q-NeuS consistently improves model performance and exhibits broad applicability across model scales and strong cross-
benchmark transferability.
Read the full abstract ▾
All authors Yanxi Pan, Jingwen Wu, Yan Ma, Xiang Chen, Xinyi Zeng, Xuetong Wang, Yunyan Zhang, Yue Zhang, Yu Tian, Jianfeng Pan, Hongyi Zhou
Instrumentation and Detectors
Response of GAGG(Ce) scintillators to different charged particles
Kai-Jie Zhou, Zai-Hong Yang, Hooi-Jin Ong et al. (+13 more)
Charged-particle
telescopes play an essential role in studies of
nuclear clustering, which require high-resolution detection of charged breakup fragments. GAGG(Ce) is a promising scintillator for charged-particle detection at high counting rates because of its high light yield, fast response, and non-hygroscopic nature. However, systematic data on its light-output response to different charged particles remain scarce. In this work, we present a systematic study of the scintillation light-output response of GAGG(Ce)
crystals to various charged particles and quantify the associated quenching behavior. The relationship between the scintillation light output and the deposited energy is established over a broad range of particle species and energies, providing essential experimental input for response modeling and energy calibration of GAGG(Ce)-based charged-particle detector systems.
Read the full abstract ▾
All authors Kai-Jie Zhou, Zai-Hong Yang, Hooi-Jin Ong, Qi-Te Li, Ze-Yu Du, Jia-Wei Bian, Cheng Wang, Jin-Ning Li, Satoru Terashima, Bing-Feng Lv, Xuan Wang, Zhi-Chao Zhang, Hong-Na Liu, Yu-Yang Yu, Peng-Jie Li, Ye-Lei Sun
Computer Vision
We Have Been Interpreting Diffusion Transformer Backbone Behaviors Wrong
Benyuan Meng, Longtao Huang, Yang Liu et al. (+1 more)
During a network forward call,
diffusion backbones naturally produce many intermediate activations. If we evaluate each activation against an identical downstream benchmarking protocol, we can plot an inverted-U-shaped curve as the depth profile: the activation's performance goes up first and then down as we go into deeper layers. This is common knowledge in the
diffusion model community, which inspires many applications including the widely-adopted REPA. However, the common interpretation of this depth profile may be inaccurate and incomplete: (i) The cause of this curve has long been treated as signal granularity changing from coarse to fine, but there has been little evidence in DiT backbones. (ii) The community has been mainly studying this phenomenon in a static way, without investigating how it changes across different training setups. In this work, we show a different mechanism picture for the profile, where the prediction target (e.g., $\epsilon$, $u$, $x_0$) sits at the center. The semantically meaningful signal is only an intermediate by-product for predicting the target, while the prediction target is the real goal for the backbone. Furthermore, the choice of prediction targets as part of the training setup can predictably affect the curve's shape. With this updated interpretation, many applications of this depth profile can be better grounded and predicted. As an example, we show how to predict the near-optimal placement of REPA based on the prediction target of the training setup. The code is at https://github.com/Darkbblue/dit-compare.
Read the full abstract ▾
All authors Benyuan Meng, Longtao Huang, Yang Liu, Liang Lin
Signal Processing
UAV SAR Imaging via Diffusion Posterior Sampling under Non-Ideal Motion in ISAC Systems
Keying Zhu, Le Liang, Jie Yang et al. (+2 more)
Unmanned aerial vehicle-enabled synthetic aperture radar (SAR) offers significant potential for providing communication services to ground users while sensing the surrounding environment in low-altitude integrated sensing and communication (
ISAC) systems. However, its practical deployment is challenged by known non-ULM trajectories and unknown position errors, which violate conventional SAR assumptions and cause severe image defocusing. We address these non-ideal impacts by formulating the
ISAC SAR imaging as an exact linear inverse problem to accurately model arbitrary trajectories with multiple-input multiple-output orthogonal frequency division multiplexing waveforms. Based on this formulation, we develop a
diffusion posterior sampling (DPS)-based super-resolution SAR framework, where a pre-trained
diffusion model provides a data-driven image prior and the exact physical model enforces measurement consistency during posterior inference. The inference process is further accelerated for practical implementation. For unknown position errors, we develop a joint iterative DPS and motion compensation (MoCo) optimization
algorithm with discrete cosine transform reparameterization and coarse-to-fine spectral scheduling. Simulations demonstrate that our framework achieves high-accuracy reconstruction across diverse trajectories and remarkable super-resolution, maintaining greater than 21.3 dB peak signal-to-noise ratio (PSNR) even under half physical resolution truncation. Under unknown motion errors, the joint DPS and MoCo
algorithm achieves 23.65 dB PSNR with a residual position-error estimation error of 0.19 cm.
Read the full abstract ▾
All authors Keying Zhu, Le Liang, Jie Yang, Fan Liu, Shi Jin
Multiagent Systems
SEPM: Self-Evolving Procedural Memory Grounded in Task Graphs for Multi-Agent Collaboration
Yi Liu, Junwei Zhao, Li Sun et al. (+2 more)
Long-horizon multi-
agent collaboration requires
memory that preserves both successful actions and the coordination structure underlying them. Existing
memory systems support information sharing or procedural reuse, but often leave the
evolution of shared procedures disconnected from the task dependencies and execution states of collaborating
agents. This coupling also limits the direct transfer of procedural
memory designed for individual
agents to team settings. We propose Self-Evolving Procedural
Memory (SEPM), a framework that uses task dependency graphs as a common representation for online coordination and procedural
memory evolution. During execution, a lightweight shared blackboard records structured evidence alongside task dependencies, supporting the detection of goal, plan, and world-state divergence and evidence-based realignment. After execution, verified actions and dependencies are distilled into reusable procedures, while recurring failures inform warnings and constraints. The shared procedure pool evolves through creation, refinement, specialization, and retirement, guided by success and cost utility and constrained by evidence and safety requirements. Together, these mechanisms preserve the collaborative structure and applicability conditions needed to reuse experience across tasks. Evaluations on ALFWorld, WebArena, and OfficeBench under AutoGen and DyLAN show that SEPM achieves the highest task scores among the compared methods in all six
benchmark settings. Under AutoGen, it improves task scores over the no-added-
memory baseline by 29.85, 25.86, and 13.00 percentage points, respectively. The codebase is available at https://github.com/cookieApril/SEPM.
Read the full abstract ▾
All authors Yi Liu, Junwei Zhao, Li Sun, Jiansong Zhang, Sen Su
Computer Vision
Compositional Concepts Erasing: Disentangling Object-Style via Domain Manifold Unlearning
Yang Xiao, Kaiyuan Deng, Gen Li et al. (+4 more)
Diffusion models pose significant risks regarding the generation of illicit content and copyright violations. This has catalyzed the development of machine unlearning to erase harmful concepts or copyrighted styles from a model's generative repertoire without compromising its general performance. However, previous unlearning approaches on
diffusion models have largely focused on either object-level or style-level removal and failed to address the challenge of object-style combination unlearning. At the same time, multiple unlearning requests often lead to catastrophic forgetting. To address these challenges, we propose Domain
Manifold Unlearning (DMU), which uses an EMA-updated feature proxy extracted from intermediate
diffusion features to represent the retained target-object direction. As a lightweight estimating representation of an object, DMU enables selective unlearning of domain-specific components in a specific object while preserving other semantic representations, thereby significantly enhancing robustness in multiple combination unlearning scenarios. Extensive experiments show that DMU significantly mitigates catastrophic forgetting, outperforming state-of-the-art methods of batch unlearning and sequential unlearning.
Read the full abstract ▾
All authors Yang Xiao, Kaiyuan Deng, Gen Li, Zinan Ling, Zhiwen Fan, Xiaolong Ma, Bo Hui
Computer Vision
TartanMatch: Towards Universal Dense Correspondence Across Modalities
Hyeokjoon Kwon, Jiting Cai, Ruogu Li et al. (+7 more)
Robotic platforms combine multi-modal sensors to operate across diverse conditions, but most dense correspondence methods are limited to RGB-RGB or a small set of modality pairs. Training a model for each modality pair scales poorly with the number of sensors and forgoes potential positive transfer across modality pairs. We introduce TartanMatch, a unified model for dense correspondence across all 25 ordered pairs among RGB, event, thermal, depth, and projected
LiDAR. TartanMatch adapts an RGB-pretrained correspondence model to all modality pairs within one shared network using lightweight modality adaptation and a mixture of real and synthetic data. On zero-shot real-world
benchmarks, TartanMatch reduces Endpoint Error (EPE) over the strongest baselines by 61.9% for cross-modal matching and 49.6% for same-modal matching. We further show that joint training across all modality pairs improves average matching accuracy over pair-specific training, especially on challenging cross-modal pairs. These results demonstrate that a single model achieves broad multi-modal coverage and state-of-the-art matching accuracy.
Read the full abstract ▾
All authors Hyeokjoon Kwon, Jiting Cai, Ruogu Li, Kritan Bhandari, Geethika Hemkumar, Parv Maheshwari, Yuheng Qiu, Yuchen Zhang, Sebastian Scherer, Wenshan Wang
Computer Vision
Spatiotemporal Recognition of Abdominal Wall Layers During Optical Trocar Entry
Shreeram Athreya, Andrew Melehy, Carter Prentice et al. (+4 more)
Objective. To evaluate whether spatial context and temporal modeling improve tissue recognition during optical trocar entry, where multiple abdominal-wall tissues may be visible simultaneously. Methods. We retrospectively analyzed 71 recordings with 1,516 annotated keyframes and 29,151 clinician-annotated sectors. Each frame was divided into 28 polar-grid sectors assigned to nine tissue classes. Color, neighboring-sector, and positional features were used by a gradient-boosted emission classifier. The full pipeline added learned temporal smoothing, spatial regularization, an anatomically ordered hidden Markov model, and posterior blending. Five-fold cross-validation was grouped by recording, with cross-fitting to prevent within-recording leakage. Primary evaluation used clinician-derived targets; ablations assessed classification, probability quality, spatial consistency, and peritoneum-or-deeper onset timing. Results. The complete pipeline achieved accuracy 0.646, macro-F1 0.507, and macro-AUROC 0.882, versus 0.600, 0.440, and 0.861 for emissions alone. Macro-F1 improved by 0.066 across all five folds. Label changes decreased from 66.2 to 19.5 per sector-minute, and median onset error decreased from 15.58 to 2.45 seconds. Frames with a tissue-depth-code span of at least three decreased from 53% to 26%, versus 22% for clinician annotations. Performance remained limited for preperitoneal fat and bowel, and the full model retained a 44% premature-onset proportion. Conclusions. Combining appearance, spatial context, and temporal constraints improved retrospective tissue recognition and onset localization. Tissue-specific limitations, residual timing errors, and reliance on future observations warrant external validation and prospective evaluation of causal models before clinical use.
Read the full abstract ▾
All authors Shreeram Athreya, Andrew Melehy, Carter Prentice, Michael Robinson, Corey Arnold, James Wu, William Speier
Econometrics
A Balanced Parametrization of Correlation Matrices
Peter Reinhard Hansen and Chen Tong
We introduce the balanced correlation parametrization (BCP), which maps positive-definite correlation
matrices one-to-one onto the set of symmetric hollow
matrices. A key feature of the parametrization is that both the transformation and its inverse are available in closed form. It combines marginal and partial correlations, is inverse-consistent and equivariant under variable permutations and sign changes, and preserves block structure. A novel property of BCP is that it preserves correlation sparsity as the dimension tends to infinity under suitable factor conditions, providing a basis for sparse high-dimensional correlation modeling. An empirical illustration using standardized idiosyncratic daily stock returns for 1,765 equities over 15 years shows that BCP retains the approximate sparsity pattern and magnitudes of shock correlations, and that it yields accurate positive-definite reconstructions under sparse coordinate restrictions.
Read the full abstract ▾
Information Retrieval
EVISTA: Unified Multimodal Video Search with Evidence-Aware Temporal Path Optimization
Minh-Huy Tran, Van-Thinh Vo, Minh-Khoi Nguyen et al. (+3 more)
In the era of big data, precise event
retrieval from large
video collections has become increasingly important and challenging. In particular, locating multiple events from the same
video in their temporal order is difficult when different events are described by different types of evidence, such as visual content, on-screen text, and spoken words. We develop EVISTA, a unified
multimodal video retrieval system for the Ho Chi Minh City AI Challenge 2026, which integrates visual-semantic, OCR, ASR, object, and metadata search over a shared keyframe index, and propose TEMPO-P, an evidence-aware temporal path optimization method for ordered multi-event
retrieval. Building EVISTA for this task reveals two limitations of independent event
retrieval: a relevant
video may be missed when one event is weakly retrieved, and independently retrieved event candidates may be temporally inconsistent or unable to represent events defined primarily by text or
speech. TEMPO-P addresses these limitations by aggregating candidate
videos across events and modalities, jointly scoring keyframes using visual, OCR, and ASR evidence, and selecting a compact temporally ordered path within each
video through dynamic programming. On five organizer queries and 36 additional multi-event queries, TEMPO-P matches visual-only temporal
alignment on visually described events (R@1 0.42 for both) and retrieves queries containing text-defined events, which visual-only
alignment cannot represent (R@1 0.38). TEMPO-P is integrated into EVISTA as its ordered-event
retrieval mode.
Read the full abstract ▾
All authors Minh-Huy Tran, Van-Thinh Vo, Minh-Khoi Nguyen, Duy-Tan Nguyen, Anh-Quan Nguyen-Tran, Loi Nguyen Khanh
Group Theory
From almost tight $p$-group enumeration to $2$-group domination among finite groups
Yinan Li and Youming Qiao
Let $f(n)$ denote the number of isomorphism classes of groups of order $n$. The pioneering works of Higman (Proc. Lond. Math. Soc., 1960) and Sims (Proc. Lond. Math. Soc., 1965), with improvement by Newman and Seeley, establish that \[ \frac{2}{27}m^3-\frac{4}{9}m^2+\Omega(m)\leq \log_p f(p^m)\leq \frac{2}{27}m^3+O(m^{5/2}). \] In the same paper in 1965, Sims speculated on improving the error term ($O(m^{8/3})$ at that time) in the upper bound. To show that $f(p^m)\leq \frac{2}{27}m^3+O(m^2)$ is now known as Sims'
conjecture. Our first main result proves Sims'
conjecture in a strong form as follows: \[ \log_p f(p^m)\leq \frac{2}{27}m^3-\frac{4}{9}m^2+O(m). \] Techniques developed to prove the above lead to progress on some well-known open questions in group enumeration in affirmative, including: (1) 2-group domination in finite group enumeration (Pyber, Eur. Congr. Math., 1996): Let $F(n)$ be the number of groups of order $\leq n$, and $F_2(n)$ be the number of $2$-groups of order $\leq n$. Then $$ \lim_{n\to\infty}\frac{F_2(n)}{F(n)} =1. $$ (2) A question attributed to Higman and Erdős (see Blackburn–Neumann–Venkataraman, Cambridge Tracts in Maths., 2007) when $m$ is large enough: $$ \text{for } 1<n<2^m, f(n)<f(2^m). $$ Our proof refines Sims' enumeration framework using graded Lie brackets, constraints on deeper commutator coordinates, and incidence and
Fourier estimates for alternating forms. Normalisation by changes of generating bases then determines the quadratic term in the upper bound. We also prove that all but a $p^{-\Omega(m^2)}$ fraction of groups of order $p^m$ have a central elementary abelian Frattini subgroup. Combining this concentration with the sharp enumeration bound of Bläser–Li–Qiao–Rogovskyy (arXiv, 2025) for $p$-groups of Frattini class $2$ yields the linear error term.
Read the full abstract ▾
Multimedia
PAG-Agent: Zero-Shot Personalized Music Album Generation via Aesthetic Profile Distillation
Yifeng Mao, Ziye Wang, Yutong Shen et al. (+2 more)
Large Music Models (LMMs) can synthesize high-fidelity audio and support increasingly flexible
prompt control, yet modeling the subjective preferences of an individual listener remains underexplored. We present PAG-
Agent, a zero-shot
multimodal framework that uses an explicit Personal Aesthetic Profile to bridge user preference inference and music generation control. A
multimodal LLM distills favorite-song audio and contextual user information into this structured representation, which a planning LLM translates into generation instructions for existing music synthesizers. We further introduce PA-Eval, which evaluates personalization through the Aesthetic Consistency Score (ACS) for target-user
alignment and the Aesthetic Specificity Index (ASI) for cross-user specificity. Experiments with Personal-100 on multiple MLLM and music-generation backbones show consistent improvements over generic and direct conditioning. A 50-participant follow-up study further shows a correlation $r=0.64$ between ACS and subjective preference.
Read the full abstract ▾
All authors Yifeng Mao, Ziye Wang, Yutong Shen, Feiran Liu, Wenjuan Han
Artificial Intelligence
MontageFlow: Physically Valid High Fidelity EEG Generation
Sudarshan Regmi, Nabaraj Subedi, Arvind Pillai et al. (+6 more)
Generative modeling offers a promising solution to the scarcity of labeled clinical electroencephalography (
EEG) data under strict
privacy constraints. Although recent generative models report substantial improvements in generation fidelity, we show that these gains conceal a categorical failure.
EEG is not an arbitrary multichannel signal: each channel records the voltage difference between two scalp electrodes. So, real recordings satisfy exact linear identities among their channels. Current generators invariably violate these identities, yet this has gone unnoticed. We therefore propose MontageFlow, a flow-matching model that enforces these identities by restricting its output to signals that satisfy them. We find that ensuring validity improves fidelity too. Furthermore, to retain the
EEG characteristics that oversmoothing washes out, we also condition the model on a physiological state vector encoded from the whole
EEG window. For generation, this vector is drawn from a prior fitted in latent space. Across 3 large-scale
benchmarks, our model improves TS-FID by 1.6-3.6$\times$ over the closest rival while ensuring validity.
Read the full abstract ▾
All authors Sudarshan Regmi, Nabaraj Subedi, Arvind Pillai, Yu Yvonne Wu, Yuliang Chen, Michael V. Heinz, Lisa Marsch, Nicholas C. Jacobson, Andrew Campbell
Algebraic Geometry
Characterization of $\mathbb{A}^1$-Milnor numbers of small rank
C-Y. Jean Chan, Thomas Hagedorn, Joel Louwsma et al. (+1 more)
The $\mathbb{A}^1$-Milnor number of a
polynomial map is a quadratic enrichment of the classical Milnor number, defined as the local $\mathbb{A}^1$-Brouwer degree of the gradient. We study which elements of the Grothendieck-Witt ring of nondegenerate symmetric bilinear forms arise as $\mathbb{A}^1$-Milnor numbers. Over fields of characteristic not $2$, $3$, or $5$, we show that every element of the Grothendieck-Witt ring of rank at most $7$ that is a local $\mathbb{A}^1$-Brouwer degree is also an $\mathbb{A}^1$-Milnor number. Unlike for local $\mathbb{A}^1$-Brouwer degrees, however, in general not all $\mathbb{A}^1$-Milnor numbers can be obtained from two-variable
polynomials. Over an arbitrary field of characteristic not $2$, $3$, $5$, or $7$, we characterize the elements of the Grothendieck-Witt ring of rank at most $8$ that are $\mathbb{A}^1$-Milnor numbers of two-variable
polynomials. Over subfields of $\mathbb{R}$, this implies that there are local $\mathbb{A}^1$-Brouwer degrees of ranks $4$, $6$, and $8$ that are not the $\mathbb{A}^1$-Milnor number of a two-variable
polynomial. We also prove that under a mild hypothesis one can reduce the number of variables of a
polynomial while preserving the $\mathbb{A}^1$-Milnor number up to multiplication by a unit.
Read the full abstract ▾
All authors C-Y. Jean Chan, Thomas Hagedorn, Joel Louwsma, Michael Wijaya
General Relativity and Quantum Cosmology
Post-Newtonian secular dynamics of hierarchical triples II. Canonical derivation of the 1PN $\times$ quadrupole cross terms and of the quadrupole-squared term
Hideyoshi Arakida
We derive the secular
Hamiltonian of a hierarchical triple from the 1PN ADM three-body
Hamiltonian by a two-stage Lie-Hori-Deprit transformation carried out entirely in canonical variables, and use it to settle the relativistic cross terms left open in Paper I. A bookkeeping rule enumerates all contributions to each order in the 1PN parameter $\delta$ and the hierarchy parameter $\varepsilon$. The 1PN monopole interaction reduces to a renormalization of the inner mass by its binding energy, and the order $\delta\varepsilon^{3/2}$, at which Lim and Rodriguez (LR) found a libration term, is empty. The electric and
magnetic 1PN $\times$ quadrupole cross terms agree exactly with the effective-field-theory result of Kuntz, Serra, and Trincherini; for the
magnetic term this is the first independent derivation. We give the next-to-leading order 1PN $\times$ quadrupole term in closed form; it depends on the outer argument of pericenter $\omega_2$. At quadrupole-squared order we obtain the complete secular
Hamiltonian, which generates the secular equations of Will and of Conway and Will; its $2\omega_2$ harmonic is a gauge term, removable even on the commensurability between the ZLK period and the nodal precession, by a generator given in closed form via Tremaine's fictitious time. The EIH and ADM descriptions coincide at 1PN order; the formulations differ only in the definition of the elements, for which we give the osculating-contact map. In secular integrations of LR's triples, the eccentricity enhancement with the octupole is controlled by the $\omega_2$-dependent part of the next-to-leading order term and is two to three times that of Paper I, while LR's accelerated mergers remain absent.
Read the full abstract ▾