Computer Vision
A Dataset-Centric Benchmark of Deep Learning Methods for Grape Leaf Disease Classification and Detection
Petar Canoski, Vlatko Spasev, Ivica Dimitrovski et al. (+2 more)
Grape leaf disease recognition is important for precision agriculture, enabling early diagnosis, timely intervention, and improved vineyard management. Although deep learning has achieved strong results, many studies rely on few
datasets, often acquired under controlled conditions, and may not reflect real vineyard challenges such as complex backgrounds, variable illumination, occlusion, leaf pose, disease severity, and device differences. This paper presents a
dataset-centric
benchmark of deep learning methods for grape leaf disease classification and detection. We analyze publicly available
datasets in terms of disease categories, annotation types, acquisition conditions, image characteristics, class distributions, provenance, and task suitability. Representative models are evaluated in three settings: image-level classification, region-level classification, and object detection. Classification is assessed using accuracy, while detection is evaluated using mAP@50 and mAP@50:95. Cross-
dataset experiments further examine transfer between
datasets with compatible disease categories but different visual and annotation characteristics. Results show near-saturated classification performance on several controlled or derivative
datasets, greater difficulty on heterogeneous
datasets, and substantial variation in detection performance across annotation settings. Cross-
dataset performance drops sharply, especially for object detection, indicating that shared disease labels do not necessarily define equivalent recognition tasks. The
benchmark emphasizes
dataset provenance, realistic field evaluation, annotation compatibility, and external validation for reliable vineyard disease recognition.
Read the full abstract ▾
All authors Petar Canoski, Vlatko Spasev, Ivica Dimitrovski, Ivan Kitanovski, Petre Lameski
Computation and Language
JuryProbe: An Empirical Consensus-Risk Diagnostic for Routing Reference-Free Factuality Judge Panels to Grounded Verification
Tianxin Zhou and Ruixi Lin
Panels of inexpensive LLM judges increasingly make accept-or-escalate decisions. In factuality settings, accepting a claim because several reference-free judges agree can create a hidden risk: agreement may reflect shared false-negative blind spots rather than independent evidence. We introduce JuryProbe, an empirical consensus-risk diagnostic for reference-free factuality judge panels, paired with a calibration-based routing policy. JuryProbe estimates consensus risk from a labeled calibration probe using false-negative-only (FN-only) judge correlation and false-consensus lift; when flagged high-risk, reference-free majority accepts are routed to the same judges with trusted references. On audited FEVER corruptions, reference-free panels show correlated false negatives (FN-only correlations 0.402 and 0.368; lifts 3.13x and 18.13x), while unanimous false consensus drops to zero under a trusted-reference best-case diagnostic on both minimal-pair and non-minimal-pair evidence. In flagged settings, the routed policy is by construction equivalent to grounding every reference-free majority accept (verified in 34/34 splits): improvement comes from accept-conditioned grounding, while the diagnostic determines whether to activate it. A fixed, pre-specified rule flags 8-10 of 10 splits across synthetic,
benchmark-authored, and scientific families and 0 of 10 on a negative control, where standing down avoids 28% of reference acquisitions at a 0.004 increase in false accepts. False-accept reduction persists under weak BM25
retrieval at substantial coverage cost, while stale stand-down labels require periodic recalibration. JuryProbe provides no formal risk guarantee and does not establish reliable stand-down on natural panels; its supported contribution is an empirical diagnostic of high-risk panel error dependence.
Read the full abstract ▾
Programming Languages
Symbolic Basic Block Profiling for Machine Learning Kernels
Jingyu Qiu, Rongcui Dong and Sreepathi Pai
Current basic block profiling techniques obtain the count of executions of each basic block in a program using dynamic instrumentation. These profiling counters create runtime overheads and also require the execution of the program, which, for large input sizes, can take substantial time. We propose symbolic program profiling that generates symbolic formulae for a basic block's count with inputs as the independent variables. Our technique is limited in applicability to a certain class of programs, namely machine learning (ML) kernels. We implement our technique in the LLVM compiler and evaluate it on 78 ML operators from 50 different ML models. These operators are generated by TVM, a machine learning compiler. Our symbolic profiles deliver exactly the same results as dynamic instrumentation for 73 out of 78 kernels with a median speedup of 15093x.
Read the full abstract ▾
Computer Vision · Cross-listed from eess.IV
Sparse Light Field Sampling Improves Casual 3D and 4D Reconstruction
Shamus Li, Ruiming Cao, Laura Waller et al. (+2 more)
Many consumer smartphones, stereo cameras, and light field cameras record multiple synchronized viewpoints in a single exposure event. However, novel view synthesis pipelines commonly use only a
monocular stream and rely on camera motion or learned priors to obtain angular coverage. In this paper, we ask: why do we use only one viewpoint? We analyze sensor-limited multi-view, where one sensor trades off spatial and angular resolution, and exposure-limited multi-view, where multiple sensors on one commodity device observe each event simultaneously. We introduce a new
dataset incorporating three types of commodity multi-view cameras, and evaluate sparse-view 3DGS and 4DGS baselines measuring reconstruction quality as a function of number of exposures and angle between extreme views. Our results demonstrate that using multiple cameras, even with a low baseline, significantly improves reconstruction quality in single-shot, few-shot, and casual
video settings. In addition, under a fixed sensor budget, angular sampling improves reconstruction when exposures are scarce despite lower spatial resolution. The gains are most pronounced for single-shot and dynamic scenes, where a stationary
monocular camera lacks the angular diversity to recover scene geometry and motion.
Read the full abstract ▾
All authors Shamus Li, Ruiming Cao, Laura Waller, Kristina Monakhova, Sara Fridovich-Keil
Formal Languages and Automata · Cross-listed from math.GR
A characterization of the reversibility of linear cellular automata
Alonso Castillo-Ramirez
Let $G$ be a group, let $\mathbb{K}$ be a field, and let $V$ be a $\mathbb{K}$-vector space. We prove that there exists a bijective linear cellular automaton $V^G\to V^G$ whose inverse is not a cellular automaton if and only if $G$ is not locally finite and $\dim_{\mathbb{K}}V \geq |\mathbb{K}|^{\aleph_0}$. In particular, $|\mathbb{K}|^{\aleph_0}$ is the exact dimension threshold for non-reversibility. This answers an open problem proposed by T. Ceccherini-Silberstein and M. Coornaert.
Read the full abstract ▾
Software Engineering
Testing and Evaluation of Agentic AI Systems In Military Command and Control
Ulysse Richard, Heather Frase, Sarah Cao et al. (+3 more)
Agentic AI systems are being procured for military command and control (C2) under public commitments to rigorous testing and human oversight. Whether such commitments can be discharged depends on their supporting assurance case, which requires three elements: claims specifying the conditions for acceptability, evidence bearing on those claims, and an argument connecting the two. Through a structured review of 240 documented Testing and Evaluation (T&E) practices, spanning eight evaluation dimensions and three lifecycle stages, we identify eight assumptions that established methods make about their test article, grouped into four clusters: system specifiability, stability, composability, and supervisability. Agentic properties weaken all eight assumptions. This erosion affects the argument connecting evidence to claims, not the claims or evidence themselves. As a result, test results may satisfy process requirements, but they do not warrant the inference from tested to fielded behavior. We derive ten assurance claims for the first three assumption clusters and assess whether current and emerging methods can address each, mapping operational consequences through five C2 scenarios. Supervisability is identified but not assessed here, since evidencing it depends on system stability results and human factors T&E methods beyond the present scope. The documented record does not support broad claims about system-level behavior, but narrower claims remain recoverable in principle, contingent on mature methods: bounded mission envelopes, trajectory-grounded correctness, executable runtime constraints, and characterized run-to-run variance. Part of the evidentiary burden shifts into deployment, making the determination to field a continuing act. Where evidence cannot be generated, the residual uncertainty can be governed through defined expiry conditions and assigned ownership.
Read the full abstract ▾
All authors Ulysse Richard, Heather Frase, Sarah Cao, Di Cooke, Sebastian Kwon, Adrianna Tan
Human-Computer Interaction
Disentangling Threads: Exploring the Potential of LLM-Supported Discussion Forum Analysis for Community Insight
Tony W. Li, Zhiqing Wang, Thanh-Nha Tran et al. (+2 more)
Online discussion forums enable people from diverse backgrounds to share ideas, feedback, and perspectives. These organic discussions can help researchers understand communities' collective viewpoints, but insights are often difficult to uncover given their freeform reply structure. Large
language models (LLMs) support qualitative text analysis but can misalign with researchers' analytical intent and miss key insights. To inform design considerations for forum sensemaking tools, we manually analyzed a forum discussion, synthesized an exploratory analysis framework from relevant literature, built a design probe, and interviewed 21 researchers to uncover perceived opportunities and barriers with LLM representations of collective discussions. We provide recommendations for community sensemaking tools to support flexible analytical goals grounded in raw user data and enable follow-up research processes, while balancing anonymous free expression with the desire for contextual information on commenters.
Read the full abstract ▾
All authors Tony W. Li, Zhiqing Wang, Thanh-Nha Tran, Yu-Chun Grace Yen, Steven P. Dow
Computer Vision
Aggregate, Don't Adapt: Subject-Level Posterior Aggregation and Transductive Calibration for Cross-Site Parkinsonian Gait Severity
Junlong Shen
We describe the winning entry to the MoCha 2026
Benchmark and Challenge on Parkinsonian Gait, which predicts MDS-UPDRS gait severity from canonicalized SMPL motion recorded at clinical sites unseen during training. The system reaches 0.6945 macro-F1 on the hidden test and ranked first of 58 entries, ahead of the runner-up at 0.5807 and the organizers' baseline at 0.4289, on a frozen public motion encoder with a single $4\times512$ linear layer. Nearly all of the margin comes from three stages usually treated as bookkeeping: reproducing the reference
benchmark's exact head recipe, averaging per-walk posteriors within the subject grouping the organizers ship, and a label-free transductive calibration of the feature mean and the decision operating point. Fine-tuning the encoder lost in four distinct forms, and ten alternative encoders were worse. Every ablation number is a paid read on the hidden test, because our own leave-two-cohort-out cross-validation proved anti-correlated with the deciding score over eleven configurations. We give the negative record in full, and identify our largest gain, subject-level aggregation, as the binding ceiling on this
benchmark.
Read the full abstract ▾
Programming Languages
Portability of Fortran's 'do concurrent' on GPUs II
Ronald M. Caplan, Miko M. Stulajter, Jon A. Linker et al. (+6 more)
There continues to be growing interest in using standard
language constructs for parallel and accelerated HPC computing, avoiding the need for (sometimes vendor-specific) external APIs. For Fortran applications,
language features such as 'do concurrent' loops open the door for compilers to implement multi-threaded, GPU-accelerated, and even distributed multi-node code with only the standard
language. Here, we explore the current status of using 'do concurrent' for GPU-accelerated Fortran applications across three major GPU vendors (NVIDIA, AMD, and Intel). Using a production application, we test their current capabilities, showing where the standard
language alone can be used, and where augmenting the code with a directive-based API (e.g., OpenMP) is still desirable or required. Multi-GPU tests are performed with GPU-aware MPI libraries. We find that the three GPU vendors can now GPU-accelerate pure Fortran (zero directives), but that manual data movement directives can help with performance and compatibility. The results show that there is rapid advancement towards making GPU-accelerated scientific HPC code performance portable using the Fortran standard
language.
Read the full abstract ▾
All authors Ronald M. Caplan, Miko M. Stulajter, Jon A. Linker, Jeff Larkin, Nikolaos Tselepidis, Harald Servat, Shiquan Su, Giacomo Capodaglio, Johanna Potyka
Cryptography and Security
More Granular, Less Trust: Enforcing Intra-Process Isolation with Arm CCA in an Untrusted Management Environment
Shiqi Liu, Zhouqi Jiang, Jie Wang et al. (+4 more)
With the increasing adoption of confidential computing, security-sensitive applications are often deployed in confidential virtual machines (CVMs), which reduce reliance on third-party cloud providers. However, privilege attacks originating from the OS remain a significant threat in these environments. Existing finer-grained isolation schemes, such as SHELTER, provide process-level protection but are still vulnerable to intraprocess attacks and potential collusion between the OS and intra-process adversaries. Many current intra-process isolation techniques continue to depend on the OS to manage and enforce isolation domains, leading to a large Trusted Computing Base (TCB). This gap highlights the need for more granular, less trust-dependent confidential computing solutions. In this paper, we present CCAegis, a system that extends the Arm Confidential Compute Architecture (CCA) to enforce intra-process isolation of sensitive data and operations, safeguarding them from both intraprocess adversaries and the OS. We employ static analysis to track the flow of sensitive data and identify functions that handle such data. Permission-switching instructions are inserted at the function call and return points, adjusting permissions via the Granule Protection Table (GPT) to ensure that only designated functions can access the isolated data. Notably, CCAegis places trust solely in the Secure Monitor, which configures the GPTs and manages domain switching, thereby minimizing the TCB. We implemented CCAegis on both an official emulator and a real development board to assess its performance. Our experimental results show that CCAegis effectively isolates sensitive data and operations, with performance overheads ranging from 1.01x to 1.43x compared to the original version across real-world cryptographic workloads.
Read the full abstract ▾
All authors Shiqi Liu, Zhouqi Jiang, Jie Wang, Wei Zhou, Kun Sun, Zhaohui Chen, Yulai Xie
Cryptography and Security
MATEE: Efficiently Bridging the Semantic Gap in TrustZone via Arm Pointer Authentication
Shiqi Liu, Xiang Li, Jie Wang et al. (+2 more)
Trusted Execution Environments (TEEs) employ hardware-based isolation mechanisms to safeguard the confidentiality and integrity of sensitive code and data. One such prevalent implementation is Arm TrustZone, which partitions the system into the secure and normal (non-secure) worlds. However, this partitioning results in the secure world having very limited visibility into the operating information of the normal world, creating a semantic gap between these two worlds. Specifically, the secure world lacks an effective user identity authentication when receiving data requests from the normal world. Consequently, malicious Client Applications (CAs) in the normal world can deceive Trusted Applications (TAs) in the secure world by utilizing elaborate request parameters, compromising the sensitive data stored by other CAs. We systematically classify these Semantic Gap Vulnerabilities (SGVs) and propose a mate system for the TEE called MATEE to defend against SGVs. MATEE utilizes Arm Pointer Authentication (PA) to bind each request to the corresponding CA's identity and then verifies the identity when the CA accesses sensitive data, thereby preventing malicious request forgery. In particular, MATEE isolates sensitive data of different CAs without modifying existing CAs and TAs. Our evaluation demonstrates that MATEE successfully defends against SGVs with a minimal runtime overhead (2.19%).
Read the full abstract ▾
All authors Shiqi Liu, Xiang Li, Jie Wang, Yongpeng Gao, Jiajin Hu
Cryptography and Security
Keyed Provenance Watermarking with Complementary Lattice-Based Secure Aggregation for Federated Learning
Xinyun Liu, Zhi Lu, Yu Chen et al. (+1 more)
Federated learning (FL) is vulnerable to multi-level attacks. However, existing methods address them separately, leaving FL exposed to data leakage, unauthorized reuse, and malicious gradient manipulation. In this work, we propose an FL framework that couples keyed context-provenance watermarking with verifiable
lattice-based secure aggregation of Real-World Anchored Watermarking and
Lattice-Based Zero-Knowledge Secure Aggregation. At the data layer, we propose a Kerckhoffs-compliant scheme that utilizes Physical Anchor Metadata (PAM) to ensure data provenance. PAM is defined as a context-provenance
token derived from trusted infrastructure data (time, location, and server ID) and then subjected to a keyed HMAC-SHA-256 transformation to produce a watermark payload that cannot be generated without the client's secret key. We further design FMGAN, a GAN-based robust image watermarking framework that embeds this transformed payload using a feature fusion module and a Mamba-guided linear
attention mechanism. At the computation layer, we adopt a
lattice-based zero-knowledge secure aggregation (LZKSA) protocol that verifies key correctness, L2 norm bounds, and cosine similarity constraints over committed gradients without revealing private updates. The RLWE-based design guarantees post-
quantum security. Extensive experiments validate the complementary protection of the two layers under composite attack scenarios. To our knowledge, no prior verification workflow has jointly evaluated both layers in a hybrid, end-to-end trustworthy FL framework.
Read the full abstract ▾
All authors Xinyun Liu, Zhi Lu, Yu Chen, Ronghua Xu
Cryptography and Security
Temporal Risk on Satellites
Shiqi Liu and Kun Sun
Satellite vulnerabilities change over time as orbits shift, power margins tighten, and the space environment deteriorates. However, most cybersecurity risk frameworks still treat threats as static. In practice, the same exploit can be far more damaging during a critical maneuver than during routine operations. We propose a temporal risk assessment framework that makes time an explicit axis in satellite security analysis. It extends existing adversary behavior taxonomies with a five-dimensional temporal capability model and estimates exploitation difficulty across distinct temporal windows of a mission. Rather than producing a single risk score, the framework outputs a series of time-indexed likelihood-impact
matrices. It discretizes missions into operationally meaningful time windows and environmental bands to show when systems are most exposed. This view helps operators avoid scheduling sensitive operations in high-risk periods and align defensive resources with a threat landscape that shifts over time.
Read the full abstract ▾
Artificial Intelligence
FlavourBench: Executable Culinary Reward Maps for Language Model Evaluation and Post-Training
Josef Chen and Erim Hayretci
We introduce FlavorBench: a
benchmark for Compiling Dense Deterministic Answer Maps from a Versioned Culinary
Embeddings Model. We test 27 frontier large
language model endpoints on 534 substitution, pairing and constraining tasks for tasks that request a 3-ingredient portfolio from 8 candidates and score all 56 resulting portfolios. We conducted multiplicity-controlled paired tests on 101 of 351 model contrasts for this task-set. The largest point estimate on this task-set was achieved by Grok 4.6 at 65.1. The same rankings for this task-set were also achieved on several independently-compiled panels (using a variety of familiar metrics, task filters, etc.) and 3 public Epicure checkpoints. We present a 3-seed post-training study where LoRA SFT of a Qwen3-0.6B checkpoint on 270 optimal answers for Epicure to score on this task-set resulted in a 13.3 point gain on 84 anchor-disjoint maps (compared to format and label-matched control; 95% CI: 6.52, 20.29; p = 0.000170).
Read the full abstract ▾
Machine Learning
Faults That Fortify: CNN Adversarial Robustness via GPU Undervolting
Behnam Omidi, Ahmad Tahmasivand, Husam Alsyouri et al. (+4 more)
Convolutional Neural Networks (CNNs) face a dual challenge: vulnerability to
adversarial attacks and prohibitive training cost.
Adversarial training is effective but expensive, a burden that grows as learning shifts to the energy-constrained edge. This paper addresses both through GPU undervolting during training. Reducing supply voltage introduces
stochastic perturbations that act as implicit regularization, improving robustness while lowering power. We characterize undervolting-induced faults at the bit level, then train LeNet, VGG-6, and MobileNetV3 on MNIST and CIFAR-10 under two training regimes, standard and
adversarial, each at nominal and undervolted voltage, and evaluate all models against
adversarial attacks. In both regimes, the undervolted model consistently achieves higher
adversarial accuracy than its nominal-voltage counterpart, showing that hardware-induced faults strengthen even
adversarial training. Because dynamic power scales quadratically with supply voltage, these robustness gains arrive with substantial energy savings. GPU undervolting is therefore a readily deployable hardware-level defense requiring no algorithmic change, and opens a promising direction in which robustness and energy efficiency move together.
Read the full abstract ▾
All authors Behnam Omidi, Ahmad Tahmasivand, Husam Alsyouri, Saba Al-Sayouri, Chongzhou Fang, Ihsen Alouani, Khaled N. Khasawneh
Software Engineering · Cross-listed from cond-mat.mtrl-sci
AutoMOOSE: Use Case and Logical Views of Agentic Phase-Field Simulation Software
Sukriti Manna, Henry Chan and Subramanian Sankaranarayanan
AutoMOOSE is an agentic software framework that converts a natural-
language request into an executed, screened, and interpreted MOOSE phase-field simulation. Here, we deploy AutoMOOSE as a agentic software, complementing our prior work which focused on development of the agentic tool. We describe our software framework and architecture through Use Case and logical views of the 1+5 architectural-views model, covering its user roles, component structure, six-
agent pipeline, physics plugin layer, Model Context Protocol interface, and screening/falsification/recovery loop. Our architecture separates physical falsification from automatic repair, so corrected simulations remain inspectable and must be re-admitted before acceptance. We focus on software design, extensibility, interoperability, and reuse of the AutoMoose framework for broad utilization in multiphysics materials design problems.
Read the full abstract ▾
Artificial Intelligence
Open-Weight Masked Introspection: Measuring What Language Models Can Report About Their Own Computation
Emilio Ferrara
Are frontier models able to introspect about their internal states? Recent work suggests that under certain conditions a complex enough model can audit its own internals, call out what changed, and report back confidently about it. We tested that claim on eight open-weight models from seven families and found no such ability: asked whether their own computation had been altered, none answered better than chance. To test it we built Open-Weight Masked Introspection (OWMI), a framework that intervenes on residual-stream sites,
attention heads and sparse-autoencoder features, then interrogates the model about the change against the null conditions an answer has to beat: sham runs where nothing was altered, impact-matched random perturbations, and a text-only observer that sees only the visible output. Over 78,000 measurements, no model's report discriminates a real intervention from a sham beyond chance (AUROC ~0.5007), and an equivalence test bounds the effect below 0.15 percentage points of AUROC. Surprisingly, all the information needed is in the models. A model fine-tuned to report this class of intervention reaches near-perfect recovery on held-out directions, and a linear probe recovers intervention presence from the same activations at 75% to 95.8% accuracy, sharpening to no held-out error at the last layer before the model speaks. In one model the signal surfaces in the confidence rather than the words: its yes-or-no report never varies, while the confidence attached to it separates intervention from sham at AUROC 0.647. The failure sits in the path from internal state to verbal report, so oversight that reads a model's own testimony needs validating against an internal reference. While our results show the inability of current open-weight models to introspect, the debate is not settled for future models.
Read the full abstract ▾
Machine Learning
AgentDecarbonizer: Carbon-Aware Execution for AI Agents
Leyi Yan, Shuangning Li and Sihang Liu
AI
agents extend large
language models from single
prompt-response interactions to long-running, goaldirected workflows that issue many model calls, invoke tools, and interact with external environments. These workflows enable tasks such as software repair, data analysis, and experiment management, but their repeated model invocations can incur substantial carbon emissions. This paper characterizes the carbon emissions of OpenClaw
agent workloads using WildClawBench, and shows that emissions depend on
token consumption, context cache reuse, and the carbon intensity of the grid. Our characterization identifies deadline flexibility as an opportunity for carbon-aware execution:
agent tasks can wait for lower-carbon-intensity periods or shift to lower-carbon grids. However, doing so requires handling uncertain execution time for temporal shifting and cached context recomputation during spatial shifting. We present AgentDecarbonizer, a carbon optimizer for AI
agents that runs alongside OpenClaw. Given a task
prompt and user-specified deadline, AgentDecarbonizer conservatively estimates task duration and selects deadline-feasible execution schedules, while accounting for cache recomputation overhead during spatial shifting. Evaluated on WildClawBench workloads with 60
agent tasks across four grids, AgentDecarbonizer reduces carbon emissions by up to 57.9 % compared with a carbon-agnostic baseline and by up to 37.5 % compared with a baseline that selects the carbon-optimal grid at task start time.
Read the full abstract ▾
Artificial Intelligence
Consilience: Conformally Calibrated Communication Control for Hidden-Profile Multi-Agent Reasoning
Abhijith Babu, Ramneet Kaur, Vishal Pramanik et al. (+7 more)
Multi-
agent LLM systems can improve
reasoning by pooling diverse perspectives, but their effectiveness depends on coordinating communication, particularly in hidden-profile settings where each
agent holds only part of the evidence required for a correct decision. Existing protocols, including fixed schedules, round-robin exchange, and unstructured debate, provide no guarantee that a conversational action is appropriate. We propose Consilience, an inference-time orchestration framework that both steers and certifies multi-
agent communication under distributed private information. At each turn, Consilience summarizes the discussion using a compact state capturing uncertainty, disagreement, evidence gain, redundancy, and premature consensus, then selects both a communication intervention (challenge, clarify, seek evidence, or route) and an appropriate speaker. Its central contribution is a round-wise conformal calibration procedure that provides a distribution-free, finite-sample guarantee: at each discussion round, conditional on reaching that round, the one-step regret of a controller's proposed action is bounded by a calibrated threshold with marginal probability at least 1 - alpha; an acceptance mechanism enforces the same guarantee for the executed action by replacing inadmissible proposals. On HiddenBench-style hidden-profile tasks spanning 12 open and closed weight
language models, Consilience improves decision accuracy and communication efficiency over fixed and unstructured discussion protocols, sometimes surpassing a full-information baseline where every
agent observes all evidence. These results demonstrate that certified adaptive communication control can be more valuable than increasing information availability, providing a practical mechanism for reliable multi-
agent LLM coordination.
Read the full abstract ▾
All authors Abhijith Babu, Ramneet Kaur, Vishal Pramanik, Olivera Kotevska, Nathaniel D. Bastian, Susmit Jha, Sunny Raj, Yanzhao Wu, Sumit Kumar Jha, Anirban Roy
Cryptography and Security
Beyond End-to-End Success: Diagnosing Failures in Long-Horizon Security LLM Agents
Wei Shao, Chongzhou Fang, Zuxiong Tan et al. (+4 more)
Long-horizon security LLM
agents must carry information and decisions across many dependent interactions, where later actions often depend on services, state, or access discovered much earlier. This makes final task success difficult to interpret: an
agent may fail before it ever reaches the point where the capability of interest can be exercised. We present a diagnostic methodology that instruments security tasks with checkpoints, separates failures before and after capability exposure, and uses controlled interventions to test suspected upstream bottlenecks. We evaluate the methodology across four task families involving delayed reuse of discovered information, reuse of observed state, recovery from failed strategies, and decision making after uncertain outcomes. On observed state reuse, checkpoint analysis shows that many Gemini 2.5 Flash failures occur before the model observes the state it is later expected to reuse. In a pre-specified 92-seed study, targeted protocol-disambiguation guidance increases state observation from 65.5% under a matched non-guidance control message to 95.4%. Repeating the same design with Gemini 3.7 Flash produces the opposite effect, while state observation no longer reliably predicts task completion. These results show that the dominant source of failure can shift across model generations, motivating evaluation that diagnoses where and why long-horizon security
agents fail rather than relying only on aggregate task success.
Read the full abstract ▾
All authors Wei Shao, Chongzhou Fang, Zuxiong Tan, Zequan Liang, Setareh Rafatirad, Avesta Sasan, Houman Homayoun
Machine Learning · Cross-listed from stat.ME
Conditional-Independence-Regularized Distributional Autoencoders for Mixed-Type Data
Siyuan Tang, Gongjun Xu and Ji Zhu
Mixed-type data containing both numerical and categorical variables arise in many scientific and real-world applications. Existing representation learning and generative modeling approaches typically focus either on reconstruction accuracy or unconditional data generation, but often fail to recover the full conditional distribution of the data while preserving interpretable structural relationships between heterogeneous variable types. In this work, we introduce Conditional-Independence-Regularized Distributional Autoencoders, a framework for learning low-dimensional representations of mixed-type data through conditional distribution matching and structural regularization. Our method combines an energy-score-based objective for numerical variables, a likelihood-based objective for categorical variables, and an auxiliary conditional independence regularization term encouraging the learned representation to capture the dependence between numerical and categorical components. We provide theoretical analysis showing that the optimal representation balances unexplained numerical variability, conditional
entropy of categorical variables, and residual conditional dependence. Empirically, the proposed method achieves strong performance on both synthetic and real-world
datasets, substantially improving categorical distribution recovery, achieving competitive overall conditional distribution recovery, and preserving mixed-type dependence structure. The code has been made available at GitHub.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from eess.IV
Consistency Models for Fast MRI Reconstruction Using Regularization by Denoising
Merve Gülle, Junno Yun, Yaşar Utku Alçalar et al. (+1 more)
Diffusion models (DMs) have emerged as powerful generative priors for MRI reconstruction with promising results. Yet DM-based methods require extensive iterative refinement, limiting their practical deployment. Consistency models (CMs) provide a compelling alternative, aiming to map out the
diffusion trajectory in a single pass, enabling faster generation. In this work, we propose CM-RED, a novel MRI reconstruction method that integrates a pretrained CM into the regularization by denoising (RED) scheme. Our method builds on accelerated proximal gradient RED (RED-APG), and further incorporates controlled noise injection during the update steps to enhance generative diversity and accelerate convergence. Extensive experiments on the fastMRI knee and
brain datasets demonstrate that CM-RED achieves high-quality reconstructions across multiple anatomies, contrast weights, acceleration factors, and undersampling patterns, using only 4 network function evaluations (NFEs). The proposed method consistently outperforms existing DM- and CM-based approaches in both quantitative metrics and visual fidelity, and exhibits strong robustness to hyperparameter variations, highlighting CM-RED as an efficient and effective generative framework for accelerated MRI reconstruction. The source code and pretrained models are publicly available at https://github.com/MerveGulle/CM-RED.
Read the full abstract ▾
All authors Merve Gülle, Junno Yun, Yaşar Utku Alçalar, Mehmet Akçakaya
Data Structures and Algorithms
Sublinear Algorithms for Estimating the Number of Hyperedges in Arbitrary Hypergraphs
Deeparnab Chakrabarty, Cooper LaPorte and C. Seshadhri
We study the problem of estimating the number of hyperedges in an arbitrary $n$-vertex hypergraph using sublinear in $n$ queries. Note that the number of hyperedges, $m$, can be exponential in $n$. For $k$-uniform hypergraphs, estimating $m$ is equivalent to estimating the average vertex degree, a problem studied in Barhum's Master's thesis (Weizmann Inst., 2007) under the standard access model of sampling random vertices, querying vertex degrees, and accessing incident hyperedges. Barhum's techniques do not extend to arbitrary hypergraphs, and simple lower-bound examples show that the standard access model cannot yield strongly sublinear
algorithms when hyperedges have unbounded size. To obtain non-trivial sublinear bounds, we consider a natural generalization of the access model called the dual access model, which allows sampling (labels of) random hyperedges, querying edge sizes, and accessing vertices in a hyperedge. In this model, we give a randomized
algorithm that returns a $(1+\varepsilon)$-approximation to $m$ with high probability, making $O(\varepsilon^{-2}\sqrt{n} + \sqrt{n}\log n)$ queries. Complementing our
algorithm, we prove a nearly matching lower bound showing that $\Omega(\sqrt{n})$ queries are necessary for any
algorithm that obtains a constant factor approximation to $m$.
Read the full abstract ▾
Computer Vision
Zero-Shot Color Image Manipulation Localization via Noise Residual Artifact Pattern Analysis
Edgar Gonzalez-Fernandez
Digital cameras embed device-specific artifacts into every acquired image through demosaicing, in-camera post-processing, and lossy compression. These traces constitute a forensic signal that can be exploited to assess image authenticity. Existing passive methods rely predominantly on the green channel of the Bayer residual, discarding the correlated information available in the remaining color channels and typically requiring training data or device enrollment. This work proposes a zero-shot, training-free blind image manipulation localization pipeline that estimates a reference artifact pattern directly from the noise residual of a single suspect image, without assuming a fixed filter configuration, color layout, or block period. The pipeline incorporates a principled denoiser selection criterion based on the acquired-to-interpolated noise variance ratio, a block-level correlation analysis against the estimated reference pattern, and a two-component Gaussian Mixture Model scoring stage that produces a pixel-level tampering probability map. An ablation study evaluates the impact of denoiser choice and block size on localization accuracy, and comparisons against state-of-the-art passive methods demonstrate the competitiveness of the proposed zero-shot approach.
Read the full abstract ▾
Computer Vision
Learning Prostate Anatomy at Test Time for Cancer Detection in Micro-Ultrasound
Obed Korshie Dzikunu, Mohammad Mahdi Abootorabi, Mohamed Harmanani et al. (+7 more)
Domain shift across clinical centers using different imaging hardware or acquisition protocols remains a fundamental barrier to deploying deep learning models for prostate cancer (PCa) detection. Existing test-time adaptation (TTA) methods address distribution shift through
entropy minimization or augmentation-based self-supervision, correcting for statistical differences in image appearance but ignoring the anatomical structure of the target domain. We propose ANT, a
segmentation-guided TTA framework that adapts a pretrained cancer detection encoder to the target domain by solving an auxiliary prostate
segmentation task at test time, supervised by pseudo-masks from a frozen pretrained
segmentation network. By aligning encoder representations to prostate anatomy in the target domain, ANT corrects domain-specific feature drift while preserving cancer-discriminative structure. The model was trained on 693 patients imaged with an earlier-generation micro-ultrasound scanner in a multi-center clinical trial, and evaluated on 118 patients acquired with a newer-generation system across two centers in another clinical trial. Under a leave-one-center-out protocol with identical evaluation conditions across all methods, ANT improves mean AUC by 2.9% and 3.6% at the biopsy-core and patient levels, respectively, over no adaptation, outperforming TTA baselines. Code is available at: https://github.com/ObedDzik/ant.git.
Read the full abstract ▾
All authors Obed Korshie Dzikunu, Mohammad Mahdi Abootorabi, Mohamed Harmanani, Paul F. R. Wilson, Emma Willis, Ferdinand Luger, Adam Kinnaird, Brian Wodlinger, Parvin Mousavi, Purang Abolmaesumi
Robotics
Logic-VLA: A Temporal Logic Conditioned Vision-Language-Action Model
Celina Shiyu Wang, Yiqi Zhao, Junjie Ye et al. (+2 more)
Vision-
language-action (VLA) models can follow natural-
language (NL) task instructions, but such instructions may not precisely specify safety-critical or spatiotemporal requirements on the resulting behavior. We introduce Logic-VLA, a formal-requirement-aware VLA that conditions on Signal Temporal Logic (STL) specifications supplied at inference time. Logic-VLA uses a syntax-graph-based STL encoder pre-trained to capture temporal logic semantics. Policy adaptation proceeds in two stages: STL-conditioned supervised fine-tuning on satisfying demonstrations is followed by trajectory-level preference optimization over matched satisfying-violating rollout pairs using a flow-matching surrogate for Identity Preference Optimization. This formulation improves formal requirement satisfaction while preserving the nominal NL task. We evaluate Logic-VLA in closed-loop quadcopter
navigation simulation across randomized photorealistic environments and test generalization to STL formulas unseen during training. Across the evaluation
benchmarks, Logic-VLA improves STL satisfaction rate over an STL-blind base policy by 24.8 to 40.7 percentage points (pp) while reducing nominal NL task success by at most 1.8 pp, showing that a single VLA can adapt its behavior to varying formal requirements without requiring a separate policy for each specification.
Read the full abstract ▾
All authors Celina Shiyu Wang, Yiqi Zhao, Junjie Ye, Yue Wang, Jyotirmoy V. Deshmukh
Cryptography and Security
aiXamine: Unified Black-Box Evaluation of Cross-Dimensional Trade-offs in LLM Safety, Security, and Privacy
Fatih Deniz, Yazan Boshmaf, Dorde Popovic et al. (+1 more)
The critical failure modes in deployed large
language models (LLMs) are cross-dimensional: a model can score 99.3 in safety
alignment while refusing one in three benign queries, or improve across every capability metric while losing 21 points in
privacy. Existing evaluation frameworks that assess safety, security, and
privacy independently cannot detect these patterns. We introduce aiXamine, a unified black-box platform that evaluates LLM trustworthiness across safety, security, and
privacy as interdependent properties. aiXamine orchestrates 46 tests across nine services through an automated red-teaming pipeline, producing hierarchical risk profiles, from
prompt-level diagnostics to cross-service trade-off analytics, that enable reproducible comparison of proprietary and open-weight systems under identical conditions. Applying aiXamine to over 120 LLMs through more than 5,000 test runs, we conduct the largest joint safety, security, and
privacy study to date and uncover three cross-dimensional phenomena invisible to single-axis evaluation. First, safety enforcement incurs a quantifiable safety tax: stronger
alignment systematically increases over-refusal, forcing providers to choose between protection and utility. Second,
privacy is near-orthogonal to other trustworthiness dimensions and not captured by standard
alignment. Third, we identify and formally characterize
distillation-induced robustness collapse: off-policy
distillation without on-policy correction causes
entropy collapse, catastrophically destroying robustness (56.9$\to$2.6) on the same base architecture. These findings, compounded by diminishing returns from scale and category-dependent safety behaviors, demonstrate that trustworthiness is inherently multi-dimensional: progress along one axis does not guarantee, and can actively undermine, progress along others, yet current
alignment methods treat it as a single objective.
Read the full abstract ▾
All authors Fatih Deniz, Yazan Boshmaf, Dorde Popovic, Issa Khalil
Information Theory · Cross-listed from math.CO
Four-Entropic Matroids Are Quaternary
Mohammad Hossein Kalantari and Shahram Khazaei
For an integer $q\ge2$, a matroid is $q$-entropic if its rank function, multiplied by $\log q$, is the joint-
entropy function of random variables on a $q$-element alphabet. We prove that a matroid is $4$-entropic if and only if it is representable over $\F_4$. The corresponding statements for alphabet sizes two and three were known. The proof combines minor closure and the excluded-minor characterization of quaternary matroids with structural properties of quasigroups of order four. Thus arbitrary four-symbol partition representations yield no matroids beyond the quaternary ones. As an application, every access structure admitting an ideal perfect scheme with a uniform four-symbol secret and four-symbol active shares also admits an ideal $\F_4$-linear scheme.
Read the full abstract ▾
Systems and Control
Wireless Power Transfer in Titanium Implants
R. W. Porto, L. Murliky, F. R. de Sousa et al. (+3 more)
This work addresses the problem of applications that require wireless power transfer to devices embedded within conductive blocks. The emergence of eddy currents tends to generate a
magnetic field opposing that of the transmitter, drastically reducing the coupling coefficient and, consequently, the efficiency of the process. A case study involving a titanium implant is presented and analyzed with a load designed to ensure a constant current of 1 mA. An experimental setup is assembled and evaluated to compare differences in the measured electrical parameters when the receiver is embedded in the implant, considering several solutions that mitigate or avoid the effects of eddy currents. The results show that such a system may be technically unfeasible if careful
attention is not paid to the geometry of the cavity in which the receiver is positioned.
Read the full abstract ▾
All authors R. W. Porto, L. Murliky, F. R. de Sousa, A. S. de Almeida, H. M. de Albuquerque, V. J. Brusamarello
Artificial Intelligence
Volumetric Radiology AI in the Era of Multimodal Large Language Models
Zanting Ye, Shengyuan Liu, Xin Liu et al. (+16 more)
Advances in
multimodal large
language models (MLLMs) are extending radiological artificial intelligence (AI) beyond task-specific image analysis toward
multimodal understanding and
reasoning. Volumetric radiology, however, presents a fundamental representational mismatch: clinical interpretation often requires full-volume spatial context and acquisition-dependent quantitative information, whereas current MLLMs are commonly conditioned on selected two-dimensional (2D) images, compressed visual representations, or report-derived text. Reliable volumetric radiology AI therefore requires representations that preserve task-relevant three-dimensional (3D) information and systems that can access, verify, and integrate this information across clinical workflows. In this Review, we examine more than 200 publications through July 2026. We organize the literature around volumetric representation and
multimodal understanding at the model level, agentic orchestration at the system level, and their links to clinical applications and evaluation. We review volumetric foundation models,
language alignment and compression strategies, and agentic systems that extend MLLMs through planning, tools,
memory, and workflow interaction. We distinguish settings in which selected 2D views or report-mediated
reasoning may suffice from those that warrant native volumetric modeling. We also introduce a Claim-Design-Validation framework to assess whether technical, workflow, and clinical claims are matched by appropriate design and validation. Across the literature, native volumetric modeling and agentic capabilities depend on the spatial, quantitative, contextual, and workflow requirements of the intended task. Clinical credibility requires faithful volumetric representation, traceable system behavior, claim-aligned validation, and clearly defined human oversight in realistic workflows.
Read the full abstract ▾
All authors Zanting Ye, Shengyuan Liu, Xin Liu, Chenhui Wang, Zhisong Wang, Jiashuai Liu, Zipei Wang, Cheng Wang, Wentao Pan, Mengjie Fang, Di Dong, Mohammad Salmanpour, Arman Rahmim, Yu Gu, Yong Xia, Hongming Shan, Yixuan Yuan, Yefeng Zheng, Lijun Lu