Discrete Mathematics · Cross-listed from math.CO
Improved Bounds for the Bilu-Linial Conjecture via Spectral Recovery from Mixed Determinantal Polynomials
Fangfang Lin and Hong Zhou
Bilu and Linial conjectured that every finite $d$-regular graph with $d \geq 2$ admits an edge signing $\sigma$ whose signed adjacency
matrix $A_\sigma$ has spectral radius at most $2\sqrt{d-1}$. We prove that every signing meeting the mixed-root condition $r_{A_\sigma}\leq\sqrt{2(d-1)}$ satisfies \[ \rho(A_\sigma) < \frac{3+\sqrt5}{2}\sqrt{d-1}, \] where $r_{A_\sigma}$ is the largest root of the mixed determinantal
polynomial $\chi[A_\sigma,-A_\sigma]$. The interlacing
theorem of Ravichandran and Srivastava guarantees a signing satisfying the mixed-root condition, so our result improves the coefficient $2\sqrt2$ in their two-sided spectral bound. In the proof, we construct a positive
matrix-valued probability measure supported on the roots of $\chi[A_{\sigma},-A_{\sigma}]$. The second moment gives a simple
matrix inequality $A_{\sigma}^2 + dI \preceq 4r_{A_\sigma}^2I$, which yields a preliminary coefficient $\sqrt{7}$. Estimates for the fourth moment use information about short walks to obtain the coefficient $(3+\sqrt{5})/2$. For triangle-free graphs, the coefficient improves to $\sqrt6$, and for graphs with girth at least five, it improves to $\sqrt{35/6}$. As a result of independent interest, we extend the construction to $\chi[A_1,\ldots,A_k]$ for Hermitian
matrices $A_1,\ldots,A_k$ with zero diagonal, and compute the first two moments explicitly. Finally, an explicit signing of $K_8$ shows that the mixed-root condition alone cannot guarantee a coefficient below $(4+\sqrt5)/\sqrt6$.
Read the full abstract ▾
Machine Learning
Knowledge-Enriched Structured EHR Features for 30-Day Hospital Readmission Prediction on MIMIC-IV
Mohamad Najafi, Hongyun Fu, Mathias Brochhausen et al. (+2 more)
Recent approaches to 30-day hospital readmission prediction rely on pre-trained
language models applied to discharge summaries. Although these methods achieve strong performance, they depend on the availability of clinical notes, incur substantial computational costs, and yield representations that lack interpretability. We propose a knowledge-enriched feature representation that augments structured Electronic Health Record (EHR) data with four medical knowledge sources: disease ontology mapping, procedure classification, drug ingredient vocabulary, and organ system laboratory aggregation, without using clinical notes. Each feature dimension corresponds to a named clinical concept, yielding a sparse and interpretable patient representation. The approach is evaluated with six classifiers on a MIMIC-IV v2.2 cohort. Under 20-fold cross-validation, the best configuration achieves an AUROC of 0.743. This performance is comparable to that of previously reported methods on this
dataset, including both those using only structured data and those incorporating clinical notes, while requiring considerably less computational cost. Interpretability analysis shows that demographics, organ system labs, drug ingredient features, and first-level ontology disease categories drive prediction, while deeper hierarchy levels contribute negligibly. These findings indicate that knowledge-enriched structured features offer a competitive and efficient alternative to
embeddings from clinical notes for 30-day readmission prediction.
Read the full abstract ▾
All authors Mohamad Najafi, Hongyun Fu, Mathias Brochhausen, Jian Wu, Yaohang Li
Systems and Control
Model-Free PID Tuning by Step-Response Inspection
Daniel Pachner, Pavel Otta, Jiří Dostál et al. (+1 more)
Industrial PID loops are still tuned by hand: step the setpoint, look at the response, change a gain. This paper turns that procedure into an
algorithm,
SPIN, for Step-response Phase-portrait INspection. From a single closed-loop step response we form three phase portraits of the control deviation and count how many times each trajectory winds around its settling point. The counts are dimensionless and independent of the step size and of the clock. They show which channel is under-damped: integral, proportional, or derivative. A triangular rule reads the three counts, cutting the gain of the one band whose fault is certain and raising that of every band the same test shows to be within its limit. No model is identified and no optimization is solved. The iteration is deterministic and its gains stay inside a fixed box. Because a ringing count is itself the evidence that the loop is in the regime where cutting that band's gain reduces the count, the iteration settles into a cycle at the edge of the well-damped set. It needs no excitation beyond a routine setpoint step, so it can run on a process in operation, and it may start from arbitrary gains, including destabilizing ones.
SPIN assumes a stable, self-regulating process whose gain and phase both fall with frequency, which covers the lag-plus-dead-time models common in process control, and a controller from the nested family $\mathrm{I} \subset \mathrm{PI} \subset \mathrm{PID}$ with integral action present. We validate it on a battery of process-typical plants with one fixed step size and one set of dimensionless constants throughout. Every decision is visible: the operator sees what the
algorithm sees. A reference implementation and an interactive demonstration are public.
Read the full abstract ▾
All authors Daniel Pachner, Pavel Otta, Jiří Dostál, Vladimír Havlena
Artificial Intelligence
Predicting build orientation for SLM dental parts: a comparison of rotation representations and direct vector regression
Felix Schmalzel, Reimar Waitz, Moritz Kronberger et al. (+1 more)
Build orientation for selective
laser melting (SLM) manufacturing of dental parts is usually chosen manually by technicians. We treat orientation prediction as supervised machine learning of the part's up-axis from technician-labeled production data, and test which rotation representations produce the best results. Using $n\approx2400$ patient-specific dental parts, we trained a ResNet-50 multi-view image backbone and a PointNeXt-S point-cloud backbone, both pretrained and fine-tuned end-to-end, on 13 up-axis representations spanning six classical $SO(3)$ parameterizations and seven representations defined directly on the unit sphere $S^2$. We report the geodesic angular error between predicted and ground-truth up-axis on a test set, with and without test-time augmentation (TTA) over $K=21$ known rotations. With TTA, the octahedral map achieves the lowest mean angular error ($10.6^\circ$, ResNet-50). The three lowest-error results overall are direct $S^2$ representations, though this may reflect label noise in the unsupervised in-plane component of the $SO(3)$ targets rather than a topological advantage. von Mises-Fisher collapses to a near-constant prediction when trained with PointNeXt-S but not with ResNet-50. TTA reduces mean angular error by 31-73 % across almost every representation and backbone. Overall, test-time augmentation over a small set of known rotations is the most consistent driver of accuracy, whereas the best-performing representation is strongly backbone-dependent.
Read the full abstract ▾
All authors Felix Schmalzel, Reimar Waitz, Moritz Kronberger, Thorsten Schöler
Information Theory
The Capacity of the Relay Channel
Jonathan Ponniah
The capacity of the relay channel is characterized by three schemes: decode-forward, coordinated compress-forward, and uncoordinated compress-forward. The proof relies entirely on properties of typical sequences.
Emerging Technologies
A Hybrid Quantum-Classical Coordination Architecture for Portfolio Optimization via Global Context Injection
Xiaoguang Yang, Menghan Dou and Guoping Guo
Near-term
quantum and
quantum-inspired solvers for portfolio optimization rely on local subproblem execution under severe size constraints, but this locality can omit cross-cluster covariance essential for global risk coordination. We propose Context-Aware Folding (CAF), a lightweight hybrid
quantum-classical coordination layer between one-shot static decomposition and full-
matrix optimization. CAF injects a compressed global risk state into each local subproblem via a state-dependent linear bias, decouples local candidate generation from global commitment, and retains a sequential acceptance rule with a monotonic non-divergence guarantee. On a 2016 Russell 3000 subset (N=484) with Simulated Annealing (SA), CAF improves the scalarized mean-variance objective by 6.59% over a static baseline (20/20 wins), and by 0.2579% on an additional 2018 panel (N=1397, 17/20 wins). We further report matched folded N=40 compatibility studies with the
Quantum Approximate Optimization
Algorithm (QAOA) and simulated
quantum annealing (SQA) as local solvers, together with a frozen hardware-in-the-loop run on Origin
Quantum's Wukong 180 (WK\_C180) for one warm-started QAOA subproblem at n=5. These results support CAF as a coordination architecture with strong classical large-scale evidence, cross-backend compatibility, and local executability on real
quantum hardware in the noisy intermediate-scale
quantum (NISQ) era.
Read the full abstract ▾
Artificial Intelligence
New Conditions for Philosophers to Catch the Wave of Citizen Deliberation in the Age of Artificial Intelligence in advance
Bernard Reber
Powerful technologies labeled “AI”-without sufficient epistemic caution-are already reshaping political and private life, bringing both new dangers and new opportunities for citizen participation. These range from electoral and legislative engagement to the most ambitious form: political co-creation through citizens' assemblies. Large
Language Models (LLMs) could support such processes through moderation, translation, facilitation, summarization, and writing assistance. But this potential remains largely unrealized. The Democratic Commons project takes a fundamentally interdisciplinary approach-from philosophy to computer science-to evaluate LLMs against five proposed democratic principles. At its core, the project is driven by the question of political bias: under what conditions can LLMs be used democratically within forms of citizen participation that are them- selves still largely experimental? Addressing these socio-technical questions requires grounding in political theory and, more broadly, in philosophy-disciplines that provide the normative frameworks without which the democratic evaluation of AI systems cannot be mean- ingfully conducted.
Read the full abstract ▾
Artificial Intelligence
Beyond Accuracy: Robustness, Cost, and Governance Trade-offs for Vision-Language Models in Templated Document Extraction
Kushal Patel, Pushkal Shrivastava, Mackenzie Lees et al. (+4 more)
Vision-
language models (VLMs) are increasingly used to extract structured fields from business documents, yet most evaluations report accuracy on clean
benchmarks and offer little guidance to practitioners choosing an approach for a given task complexity. We address this gap with a measurement-grounded study and an open-source release. Across eleven systems (three commercial, two
reasoning, five open-source VLMs in pretrained and fine-tuned form, and a non-LLM OCR->regex floor) scored on a 750-document held-out pool of synthetic checks, fine-tuning on 3K samples lifts the best open-source VLMs above F1 0.98-above every zero-shot commercial system on this task-while GPT-5 leads the commercial pool on F1 and Claude Sonnet 4.5 collapses on Date. To turn these measurements into actionable choices, we introduce a practitioner-oriented selection framework that maps a task profile (quality, latency, governance, volume) to a recommended approach via filtering and total-cost minimization, illustrated on a hypothetical mid-volume document-extraction scenario.
Read the full abstract ▾
All authors Kushal Patel, Pushkal Shrivastava, Mackenzie Lees, Qirui Lu, Bhargobjyoti Saikia, Liying Li, Junlin Jiang
Networking
AI-Native Open RAN: A Roadmap from xApps and rApps to Autonomous Network Agents
Ryan Barker, Alireza Ebrahimi Dorcheh, Tolunay Seyfi et al. (+6 more)
Open
Radio Access Networks (O-RAN) have emerged as a transformative paradigm for future wireless systems by introducing openness, virtualization, disaggregation, and programmable intelligence through the RAN Intelligent Controller (RIC). The availability of standardized interfaces and near-real-time control loops has created unprecedented opportunities for integrating artificial intelligence (AI) into
radio access network management and optimization. Over the past several years, a broad range of AI techniques have been proposed to address key O-RAN challenges such as
radio resource management, network slicing, traffic prediction, mobility management, interference mitigation, and spectrum sharing. Despite significant progress, existing solutions often remain task-specific, require extensive retraining, and exhibit limited generalization across deployment environments and network conditions. This paper presents a comprehensive review of AI-enabled O-RAN systems and provides a unifying perspective on the
evolution of intelligence in wireless networks. We first examine the O-RAN architecture and the role of intelligence within near-real-time and non-real-time RIC frameworks. We then develop a taxonomy of AI approaches for O-RAN, covering machine learning, deep reinforcement learning (DRL), digital-twin-assisted optimization, and emerging foundation-model-based architectures.
Read the full abstract ▾
All authors Ryan Barker, Alireza Ebrahimi Dorcheh, Tolunay Seyfi, Mohammad Raihan Uddin, Alireza Mohammadhosseini, Julia Boone, Stephen Streit, Drew Schlesener, Fatemeh Afghah
Machine Learning
Backward SDEs-based Diffusion for Physics-Constrained Generation
Zihao Wang
Pretrained score-based
diffusion models provide strong unconditional priors, yet enforcing measurement or physics consistency in inverse problems is often handled by heuristic guidance, intermittent projections, or task-specific conditional training, with limited guarantees of feasibility at the end of inference. We propose terminal-conditioned inversion for score-based SDE priors. Given a frozen Score-SDE prior and a task-defined terminal feasibility specification, we construct an associated backward
stochastic differential equation whose adapted solution defines a principled inverse map from the terminal requirement to a prior state at a chosen noise level. Under standard regularity conditions, we establish existence and uniqueness of the adapted solution and obtain terminal consistency by construction. We further develop a practical neural BSDE solver that composes arbitrary pretrained
diffusion priors with domain constraints without modifying the score-defined coefficients, producing an anchored prior state that enables neighborhood sampling for uncertainty characterization. Experiments on toy
datasets validate stable terminal-conditioned inversion and distributionally consistent neighborhood sampling. As a real-world case study, we apply the framework to sparse-view CT reconstruction and achieve improved reconstruction quality over representative training-free baselines while satisfying strict measurement feasibility under the prescribed terminal specification. Project is available in: \href{https://laplacelab.github.io/BSDEDiffusion/}{https://laplace.center/icmlbsdeI/}
Read the full abstract ▾
Mathematical Software · Cross-listed from quant-ph
OpenQARP: a modular framework for quantum application research
Stefano Scali, Vicente P. Soloviev, Antonio Márquez Romero et al. (+7 more)
We introduce OpenQARP, the Open
Quantum Application Research Package: an open-source Python framework for
quantum application research, built on a compiled C++ core. In OpenQARP, an application is assembled from an
algorithm library that orchestrates three interchangeable layers: blocks describe
circuits, primitives describe what to extract from them, and engines describe how they run. The release spans near-term through fault-tolerant methods, targeting problems from electronic structure to combinatorial optimization, with device-aware compilation,
circuit cutting, and resource estimation over a single set of numerical conventions that the code must follow. Every numerical feature is checked against an independent oracle, and every published timing carries a correctness check. Against OpenFermion, Qiskit/Aer, PennyLane/Lightning, Qulacs, qsim, and pytket, OpenQARP is one to two orders of magnitude faster on operator
algebra and among the fastest on simulation. It matches mature compilers on two-
qubit gate count, and expresses a complete
algorithm in the lines a framework needs rather than the plumbing a primitive stack demands.
Read the full abstract ▾
All authors Stefano Scali, Vicente P. Soloviev, Antonio Márquez Romero, Brian Coyle, Giuseppe Buonaiuto, Annie Paine, Jonathan H. Fetherolf, Marcos Diez García, Michal Krompiec, Josh Kirsopp
Human-Computer Interaction
More Than Just Access: Generative AI as Communication Intermediary for Blind and Low-Vision Users
Protik Dey, Mohd Saifuzzaman and Taslima Akter
Generative AI (GenAI) tools are increasingly woven into how blind and low-
vision (BLV) people communicate, not only with digital information, but with the physical world and with other people. Tools such as ChatGPT, Google Gemini, Be My AI, and Seeing AI translate visual and textual content into accessible form, and are beginning to substitute for interpersonal requests for help, such as asking a family member to read a label or describe a scene. Drawing on semi-structured interviews with 19 BLV participants, we examine GenAI as a communication intermediary and how it succeeds and fails as an alternative for reading, describing, and even asking another person for help. We also investigated what BLV users gain and risk when these tools take over that role. We conclude with design and policy implications for GenAI systems that communicate uncertainty honestly, protect information, and support BLV users' independence rather than substitute for it unsafely.
Read the full abstract ▾
Artificial Intelligence
NoteVQA: Benchmarking VLMs on Real-Life Questions from Human Communities
Haonan Jiang, Guojian Zhan, Jiancong Xie et al. (+5 more)
Vision-
language models (VLMs) increasingly power consumer-facing AI search, yet evaluating them on the diversity of everyday visual questions remains challenging. Existing
benchmarks often target predefined capabilities, such as multi-hop
retrieval or long-form synthesis, whereas users ask photo-grounded questions spanning a long tail of everyday scenarios. Despite advances in VLMs, users on Xiaohongshu, a mainstream Chinese image-sharing platform, continue to turn to other people for help with everyday visual questions. Motivated by this behaviour, we curate NoteVQA from these questions, yielding 252 items across 12 topical categories and 7 user intents. Each item includes a concise reference distilled from expert community responses and a human-audited interleaved reference answer that combines textual explanations with supporting visual evidence. We evaluate both short-answer correctness and interleaved-answer quality. To support the latter, we introduce AgenticInterleave, a single-
agent ReAct framework for
retrieval-supported answer generation, together with IVR-12, a 12-dimensional rubric for assessing the content, presentation, and image quality of interleaved references and model outputs. Across 9 frontier VLMs, the highest short-answer accuracy is 52.8%, while adding agentic search to Qwen3.5-397B-A17B improves accuracy by only 2.0%. For interleaved answers, the same model running AgenticInterleave scores 3.52 under IVR-12, compared with 4.65 for the human-audited references, with the largest gap in content quality. These results highlight the challenges that everyday visual questions pose for current VLMs in both answer accuracy and the quality of visually grounded explanations.
Read the full abstract ▾
All authors Haonan Jiang, Guojian Zhan, Jiancong Xie, Shijun Wan, Dongiia Zhao, Cheng Chen, Yahui Liu, Chuan Mu
Cryptography and Security
HYDRA: Quantifying Botnet Resource Thresholds for Efficient Link-Flooding Attacks on LEO Satellite Networks
Roee Idan, Rami Puzis, Asaf Shabtai et al. (+1 more)
Low Earth orbit (LEO) satellite constellations, such as Starlink and Kuiper, are rapidly emerging as a critical backbone for low-latency global connectivity. As these systems expand, they become more attractive attack targets, necessitating increased resilience and security. Threat actors seek to exploit constellation-specific properties such as predictable motion, time-varying
topologies, and reliance on inter-satellite and ground-satellite links. Recent work has shown that link-flooding attacks (LFAs) can exploit these properties to congest strategic network bottlenecks. Yet, prior work does not quantify the resilience of LEO networks to targeted disruption. We present HYDRA, a modeling and optimization framework that formulates LFA variants as botnet minimization problems. HYDRA quantifies network resilience to LFAs by measuring the smallest active subset of bots and the corresponding traffic allocation required to disrupt communication between targeted geographic areas. Under matched stealth constraints, HYDRA achieves the same targeted disruption as ICARUS while using 34% fewer bots and 23% less aggregate attack traffic. HYDRA achieves over 97% success in sustaining continuous attacks as the network
topology evolves. Finally, HYDRA evaluates five mitigation strategies, showing how routing diversification, ingress policing, distance-based traffic constraints, source throttling, and botnet attrition reduce attack success and improve network resilience to targeted disruption.
Read the full abstract ▾
All authors Roee Idan, Rami Puzis, Asaf Shabtai, Yuval Elovici
Neural and Evolutionary Computing
Complete Suffix Prediction for Recommendation via Latent Retrieval over Process Graphs
Sarra Madad, Myriam Maumy, Frédéric Bertrand et al. (+1 more)
Complete suffix prediction is challenging in sequential decision settings, where the same prefix can remain compatible with several plausible suffixes. We propose a graphbased metric-learning framework that reformulates complete suffix prediction as latent
retrieval over process graphs. Prefixes and suffixes are represented as directed attributed graphs and encoded by edge-conditioned graph neural networks, allowing event-level activities and transition-level durations to be modelled jointly. Prefix representations are projected into the latent suffix space through a predictor trained with a joint reconstruction and contrastive objective strengthened using process-aware hard negatives. To stabilise the learned
retrieval geometry, spectral normalisation, and
retrieval robustness, spectral normalisation is applied to enforce a Lipschitz constraint on both encoders and predictor. Experiments on two real-life process
datasets demonstrate that the proposed framework achieves the best overall results across nearly all evaluated criteria. It improves semantic suffix accuracy measured by normalized Damerau-Levenshtein distance, yields strong
retrieval quality through Recall@1, Recall@5, and MRR@5, and maintains temporal plausibility according to Mean Absolute Error. These results show that graph-based latent
retrieval is an effective alternative to sequential suffix prediction for recommendation-oriented process monitoring under structural and KPI-related constraints.
Read the full abstract ▾
All authors Sarra Madad, Myriam Maumy, Frédéric Bertrand, Yoann Valero
Artificial Intelligence
EEG-Xplain: Decoding Neural Black-Boxes of EEG Foundation Models
Hansong Ma and Junxiao Wang
EEG foundation models such as BIOT, LaBraM, and EEGMamba have achieved remarkable performance in neural signal decoding, but their black-box nature limits clinical trust and neuroscientific validation. We propose a unified attribution framework for interpreting
EEG foundation models across heterogeneous architectures. The framework integrates gradient-, perturbation-, and activation-based explanation methods to analyze model behavior in spatial, temporal, and frequency dimensions. Spatially, it identifies critical
EEG channels and visualizes their distributions using topographic maps. Temporally, it highlights decision-relevant signal segments through attribution heatmaps. In the frequency domain, it quantifies the contributions of canonical
EEG rhythms via spectral perturbation analysis. To assess explanation reliability, we introduce a population-level evaluation combining Area Over the Perturbation Curve (AOPC) and cross-method consistency analysis. The framework further leverages Large
Language Models (LLMs) to transform structured attribution outputs into natural-
language reports, bridging low-level neural representations and high-level semantic
reasoning. Experiments on
benchmark datasets, including Mumtaz2016 and TUAB, demonstrate that the generated explanations are consistent with established neurophysiological markers, validating meaningful neural representations while exposing potential dependencies on artifacts and spurious patterns. The proposed framework provides a standardized approach for evaluating the interpretability, reliability, and physiological plausibility of
EEG foundation models.
Read the full abstract ▾
Software Engineering
Technostress in the Age of AI: A Preliminary Study with Software Professionals
Ronnie de Souza Santos, Italo Santos, Cleyton Magalhaes et al. (+1 more)
The rapid adoption and
evolution of AI are changing software engineering work and requiring professionals to repeatedly adapt their knowledge, practices, and skills. Although technological adaptation has long characterized software development, less is known about how these new and recurring pressures manifest as technostress. This preliminary exploratory study investigates AI related technostress among software professionals. We conducted a survey and performed a thematic analysis of responses from 121 software professionals across 26 countries reporting their recent experiences with AI at work. Our findings suggest that AI related technostress emerges not only from adapting to rapidly changing technologies, but also from having to manage the work, technical responsibilities, and professional changes that accompany their adoption. This characterization shows that AI introduces pressures beyond learning and using new tools, affecting how software professionals perform and remain accountable for technical work and how they prepare for the future of their careers.
Read the full abstract ▾
All authors Ronnie de Souza Santos, Italo Santos, Cleyton Magalhaes, Zixuan Feng
Human-Computer Interaction
Can a Neural Encoding Model Replicate an fMRI Visualization Study?
Erfan Nasirzadeh Orang and Zack While
Most knowledge of graphical perception comes from behavioral studies. Understanding from a neural perspective is much more limited due in part to neuroimaging studies' expensiveness and difficulty to conduct. In this paper, we evaluate whether Meta's Tribe V2 neural encoding model can recover neural contrasts from a visualization fMRI study. Specifically, we evaluate Tribe V2 through a conceptual replication of the visualization-viewing component of a prior comparison of Bubble charts and three-dimensional Surface charts in color and grayscale. We generate TRIBE-predicted cortical responses for the original stimuli and compare the resulting contrasts with those reported in the human study. The model reproduced the direction of 11 of 14 reported cortical effects, with agreement concentrated in visual-processing regions. This agreement characterizes the model's
alignment with the prior human-generated fMRI results rather than independently confirming them. We discuss the limitations encountered when working with this model for in-silico replication and hope to encourage future work exploring this new avenue for neuroimaging studies in visualization. Supplemental materials are available at https://osf.io/8a96x/.
Read the full abstract ▾
Computation and Language
RESKILL: Explicit Failure Attribution and Structured Repair for Interactive Language Agents
Mengyi Deng, Xin Li, Duyi Pan et al. (+4 more)
Language agents increasingly rely on reusable skills, but post-failure repair is often handled by opaque one-shot reflection: a model generates a skill patch without explicitly maintaining how failure explanations relate to candidate repairs or how unsuccessful retests should influence later edits. We introduce RESKILL, a structured repair framework that maintains an explicit repair state across repair rounds. Given a failed rollout, the framework links failure hypotheses to candidate skill patches, selects local repairs through coverage-based attribution, retests the edited skill set in the environment, and uses retest outcomes to guide subsequent repair updates. The
language model supplies structured repair factors, while the repair procedure records them, compares local skill patches by how well they address active failure explanations, and carries unsuccessful retest outcomes into later repair rounds. We evaluate RESKILL on ALFWorld and TextCraft across three model sizes under fixed repair budgets. RESKILL obtains the strongest final success in all six
benchmark-model settings, improving average final success by 3.7 percentage points over direct repair and 3.3 points over hypothesis-conditioned repair. These results suggest that explicit attribution alone is insufficient; durable improvement emerges when attribution is integrated with repair selection and persistent retest-conditioned update.
Read the full abstract ▾
All authors Mengyi Deng, Xin Li, Duyi Pan, Zilin Wang, Zhiwei Li, Zhijiang Guo, Wei Wang
Computer Vision
V-ICAL Bench: Evaluating Video In-Context Learning for Multimodal Agents in Interactive Environments
Ziqian Fan, Shibo Xu, Junjie Li et al. (+8 more)
While In-Context Learning (ICL) enables models to adapt from exemplars without parameter updates,
multimodal ICL remains largely underexplored, particularly regarding
video demonstrations in interactive environments. For
multimodal agents, learning from
videos presents unique challenges: they must translate in-context demonstrations into executable policies, ground these policies in novel visual states, and iteratively refine actions based on environmental feedback. We introduce V-ICAL, a novel
benchmark designed to evaluate
video-based ICL for
multimodal agents. Comprising 342 interactive tasks across 37 environments, V-ICAL utilizes human-curated demonstration
videos as task-specific behavioral exemplars, evaluating
agents through sustained interaction from a target initialization. The
benchmark seamlessly connects in-context knowledge induction with core agentic capabilities, including state grounding, temporal
memory, planning, and adaptation in dynamic environments. Extensive evaluations across 19 state-of-the-art
multimodal agents reveal significant limitations: the best-performing model, Seed-2.1-Pro, achieves a score of only 54.4/100, while other leading models (e.g., Gemini-3.1-Pro, GPT-5.6) fail to surpass 50, far below the human baseline of 83.6. Controlled comparisons further demonstrate that current
agents struggle to reliably translate
video exemplars into effective policies, failing to yield consistent performance gains. Ultimately, V-ICAL exposes a critical gap in the ICL capabilities of
multimodal agents, underscoring an urgent need for future research.
Read the full abstract ▾
All authors Ziqian Fan, Shibo Xu, Junjie Li, Xiangyu Zhao, Shengyuan Ding, Yifan Yang, Zhenjie Yang, Haodong Duan, Yue Zhou, Zhihang Zhong, Xue Yang
Distributed Computing
Scalability and Performance Evaluation of Federated Learning Frameworks: A Comparative Analysis
Bassel Soudan, Sohail Abbas, Ahmed Kubba et al. (+2 more)
This paper presents a systematic examination and experimental comparison of the prominent
Federated Learning (FL) frameworks FedML, Flower, Substra, and OpenFL. The frameworks are evaluated experimentally by implementing
Federated Learning over a varying number of clients, emphasizing a thorough analysis of scalability and key performance metrics. The study assesses the impact of increasing client counts on total training time, loss and accuracy values, and CPU and RAM usage. Results indicate distinct performance characteristics among the frameworks, with Flower displaying an unusually high loss, FedML achieving a notably low accuracy range of 66% to 79%, and Substra demonstrating good resource efficiency, albeit with an exponential growth in total training time. Notably, OpenFL emerges as the most scalable platform, demonstrating consistent accuracy, loss, and training time across different client counts. OpenFL's stable CPU and RAM underscore its reliability in real-world scenarios. This comprehensive analysis provides valuable insights into the relative performance of FL frameworks, offering good understanding of their capabilities and providing guidance for their effective deployment across diverse user bases.
Read the full abstract ▾
All authors Bassel Soudan, Sohail Abbas, Ahmed Kubba, Manar Wasif Abu Talib, Qassim Nasir
Robotics
Volumetric Harmonic Field Navigation for Quadrotors
Shuxiu Jia, Amartya Mukherjee and Yating Yuan
Quadrotor
navigation in cluttered 3-D environments requires global guidance while local motion remains subject to collision and motion limits. Harmonic potentials provide dense guidance from a global boundary value problem, but coupling a volumetric harmonic field to constrained physical quadrotor motion remains an open experimental problem. We couple a precomputed volumetric harmonic field with a constrained predictive planner that queries the field at predicted positions instead of extracting a global reference path. In Structured 3-D tests, harmonic guidance yields larger minimum clearance and lower RMS jerk than matched Dijkstra guidance, at the cost of longer paths; the same pattern remains when both methods use the same passage. Long maze tests span routes far beyond one prediction horizon, and Crazyflie trials validate physical execution. To the best of our knowledge, this is the first physical quadrotor demonstration of volumetric harmonic field
navigation. The results show that globally constructed harmonic guidance can directly support local constrained motion generation on a physical quadrotor.
Read the full abstract ▾
Machine Learning · Cross-listed from math.OC
Projection-Free Multi-level Algorithms for Stochastic Constrained Compositional Optimization
Wei Jiang, Sifan Yang, Wenhao Yang et al. (+4 more)
This paper studies projection-free
algorithms for
stochastic constrained multi-level compositional optimization. In this context, the objective function is a nested composition of several smooth functions, and the decision set is closed and convex. Since projection onto the constraint set can be computationally expensive, we develop projection-free methods that rely on linear minimization oracles. For non-convex objectives, we propose variance-reduced projection-free
algorithms and establish complexity guarantees under both the Frank-Wolfe gap and the gradient mapping criteria. We also develop momentum-based methods that achieve convergence guarantees under weaker smoothness assumptions. Additionally, by using a stage-wise design, we derive a parameter-free variant that preserves the same complexities for the Frank-Wolfe gap. Such a design can be further used to develop
algorithms for convex and strongly convex functions whose rates match those of single-level projection-free counterparts. Finally, we consider finite-sum problems and derive complexities for non-convex, convex, and strongly convex objectives. Numerical experiments across multiple tasks demonstrate the effectiveness of the proposed methods.
Read the full abstract ▾
All authors Wei Jiang, Sifan Yang, Wenhao Yang, Yibo Wang, Yuanyu Wan, Zechao Li, Lijun Zhang
Discrete Mathematics · Cross-listed from math.CO
Certificate Complexity of Elementary Symmetric Boolean Functions of Arbitrary Degree
Jing Zhang and Yuan Li
Let $\sigma_{n,d}$ denote the elementary symmetric Boolean function of $n$ variables and degree $d$. Previous work determined $C(\sigma_{n,d})$ when $d$ is odd or a power of two, but the general even non-power-of-two case remained open. We determine the certificate complexity for every degree $1\le d\le n$, thereby completing the classification for elementary symmetric Boolean functions. Writing $d=2^t m$ with $m$ odd, we obtain an explicit formula in which the possible deficit from the maximal value $n$ is controlled by $2^t$, while the exact value is determined by a binary containment condition involving $m$. In particular, \[ n-2^{\nu_2(d)}+1\le C(\sigma_{n,d})\le n, \] and we characterize when the upper bound is attained. Moreover, we determine the least positive period of the certificate-complexity deficit $\Delta_d(n)=n-C(\sigma_{n,d})$: it is $1$ for odd $d$, equals $d$ when $d$ is a power of two, and equals $2^{\lfloor\log_2 d\rfloor+1}$ for even non-power-of-two $d$. This least-period problem is distinct from the classical periodicity of the underlying value sequence $\binom{j}{d}\bmod 2$. The known odd-degree and power-of-two formulas are recovered as special cases.
Read the full abstract ▾
Data Structures and Algorithms
Breaking the 1/3 Barrier for constrained $\boldsymbol{k}$-Submodular Maximization: A Proportional Top-2 Randomized Framework
Siyuan Chen, Shengminjie Chen, Suixiang Gao et al. (+3 more)
$k$-submodularity generalizes submodularity by allowing each selected element to be assigned one of $k$ labels. We study the maximization of a non-monotone $k$-submodular function, where $k\ge2$, under matroid and knapsack constraints. We develop a proportional top-2 randomized framework that achieves an approximation ratio of $\sqrt2-1\approx0.4142$ under either constraint, breaking the previous $1/3$ barrier. Once an element is selected, its label is chosen from the two labels with the largest marginal gains, with probabilities proportional to their positive parts. The analysis separates replacement, pure-addition, and residual-deletion losses, and proves a sharp local loss certificate for this rule. We then give a Bellman analysis using the same local certificate. It gives a $1/2$-approximation under a cardinality constraint and strengthens the knapsack guarantee to $1/2$ without changing the
algorithm. The $1/2$ guarantees are oracle-optimal for $k=2$ and asymptotically tight as $k\to\infty$. The value-oracle complexities are $O(Bnk)$ for cardinality, where $B$ is the cardinality bound, $O(n^2k)$ for matroid, and $O(n^3k^2)$ for knapsack.
Read the full abstract ▾
All authors Siyuan Chen, Shengminjie Chen, Suixiang Gao, Zheyu Jiang, Chenhao Wang, Wenguo Yang
Computer Vision
MambaMPD: A Mamba-Driven Segmentation Framework for Marine Pollution Detection from Remote Sensing Imagery
Shuaiyu Chen, Wei Han, Peng Ren et al. (+2 more)
Accurate marine pollution detection (MPD) is essential for protecting coastal ecosystems and marine biodiversity.
Vision Mamba models have shown promise in remote-sensing semantic
segmentation by efficiently capturing long-range dependencies and global context, yet their potential for MPD remains underexplored. MPD is particularly challenging because of low signal-to-noise ratios, fragmented pollution patterns, and indistinct boundaries caused by the visual similarity between pollutants and the surrounding sea. To address these issues, we propose MambaMPD, an enhanced Mamba-based framework incorporating two complementary structural priors: Frequency-Aware Augmentation (FAA) and multi-scale Edge-Guided
Attention (EGA). FAA integrates wavelet transforms into the encoder to decompose features into multi-scale frequency subbands, enabling the model to capture low-frequency contextual semantics and high-frequency structural details needed to identify small, low-contrast, and irregular pollution patterns. EGA adaptively fuses hierarchical,
Laplacian-derived boundary cues with deep semantic representations, refining encoder features before decoding to sharpen boundaries and reduce ambiguity in visually confusing, spatially fragmented scenes. Together, these modules improve sensitivity to subtle pollution signals while preserving fine boundary structures. A U-Net-style decoder with squeeze-and-excitation
attention and deep supervision progressively restores and refines semantic and spatial information across scales. Extensive experiments on two
benchmark MPD
datasets show that MambaMPD achieves higher mIoU than competing methods while requiring substantially less computation than foundation-model-based approaches. On MADOS, it improves F1 by 3.6% over OSDMamba; on M4D, it raises Oil Spill IoU by 6.82% over TransOilSeg.
Read the full abstract ▾
All authors Shuaiyu Chen, Wei Han, Peng Ren, Chunbo Luo, Zeyu Fu
Sound
Sectional Structure and Emotional Dynamics in Chinese Pop Songs: An Empirical Analysis of Valence-Arousal Trajectories across 100 Songs
Jingyi Lyu
Music
Emotion Recognition (MER) aims to identify and represent emotional information in music through computational methods and is an important research area within Music Information
Retrieval (MIR). To address the limited consideration of song sectional structure in existing dynamic MER research, this study examines 100 Chinese pop songs by aligning 1,046 manually annotated sections with continuous Valence-Arousal (VA) trajectories and analyzing them from the perspectives of section type, adjacent section transitions, repeated sections, and whole-song trajectories. The results show that emotional differences across sections are reflected more strongly in Arousal. Verse typically forms a relatively low-activation baseline, Pre-chorus exhibits a progressive buildup, and Chorus produces a more pronounced high-arousal arrival, while Interlude, Bridge, and Outro tend to show transitional, divergent, and closing functions, respectively. Although whole-song sectional configurations are diverse, high-frequency local transitions are relatively concentrated. High-arousal positions occur more often in the later part of a song, but are not fixed to the final Chorus. Based on these findings, this study summarizes the emotional organization of the selected pop songs as an empirical framework of "local cycles-global accumulation", in which local sectional cycles are accompanied by emotional pullbacks, while the overall trajectory exhibits a later-stage rise in VA and a tendency for high-Arousal positions to occur later in the song.
Read the full abstract ▾
Computer Vision
Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models
Md Khalid Syfullah and Alvi Ataur Khalil
Visual Question Answering (VQA) with
Vision-
Language Models (VLMs) is increasingly used in
privacy-sensitive and bandwidth-constrained settings.
Federated Learning (FL), Split Learning (SL), and U-Shaped Split Learning (USL) keep raw data local, but transmitting all visual
tokens across a model partition remains costly and can expose private information. We propose QPriv-VL, a question-guided,
privacy-aware
token-
pruning framework for FL, SL, and USL that prunes visual
tokens before transmission based on task utility and
privacy sensitivity. Its core component is a lightweight Dynamic Threshold Predictor (DTP) that jointly estimates a sample-specific
pruning ratio and a
token-level retention mask in one forward pass. DTP combines question relevance, computed from cross-modal similarity between visual patches and the pooled question
embedding, with a sensitivity signal derived from frozen DINOv2 features. This allows the model to suppress potentially sensitive regions while preserving patches useful for answering the question, without requiring sensitivity labels. We evaluate QPriv-VL on GQA, OK-VQA, VQAv2, SLAKE, VQA-RAD, and PathVQA against four
privacy attack families: FSHA, FORA, iDLG, and attribute-inference membership inference attacks. DTP matches or outperforms fixed-ratio
pruning while using substantially fewer transmitted
tokens. On VQA-RAD, it reduces membership-inference attack success from 0.99 to 0.76-0.79, lowers FSHA and FORA reconstruction PSNR relative to fixed-ratio
pruning, and preserves competitive VQA accuracy using about 40% of the original visual-
token budget. A sensitivity exclusion ratio of 1.20 +/- 0.18 indicates preferential removal of
privacy-sensitive patches, while explainability analysis shows that retention adapts to question semantics rather than generic visual saliency.
Read the full abstract ▾
Computer Vision
Benchmarking Intra-Patient 3D Deformable Multimodal Image Registration
Matteo Barbieri, Giammarco La Barbera, Juan Pablo De La Plata et al. (+3 more)
Multimodal image registration is a key component of many clinical workflows, yet it remains challenging because corresponding anatomical structures often exhibit substantially different image intensities across modalities. In this work, we present a comprehensive
benchmark of intra-patient 3D
multimodal deformable registration methods across three
datasets covering different anatomical regions and difficulty levels, including both synthetic deformation recovery and real clinical scenarios. We evaluate classical optimization-based approaches and modern learning-based methods, including recent deep learning and foundation models, using complementary metrics: Average Dice similarity coefficient (DSC), average 95th-percentile Hausdorff distance (HD95), and a modality-independent structural similarity measure based on the MIND self-similarity context (MIND-SSC). Results show high variability across
datasets, with learning-based methods demonstrating superior performance on large synthetic
benchmarks, while only limited improvements are observed in real pelvic registration. A key finding of this study is the consistent disagreement between
geometric metrics (DSC, HD95) and image-based similarity metrics (MIND-SSC), highlighting that improved overlap does not necessarily imply better global
multimodal correspondence. Furthermore, anatomy-guided approaches achieve the highest overlap scores but exhibit degraded performance outside of segmented regions, revealing a trade-off between label-driven
alignment and global structural coherence. Overall, our results indicate that no current method achieves robust performance across anatomies and modalities. We demonstrate that intra-patient 3D
multimodal registration requires multi-criteria evaluation, including deformation-based metrics, and remains an open problem.
Read the full abstract ▾
All authors Matteo Barbieri, Giammarco La Barbera, Juan Pablo De La Plata, Sabine Sarnacki, Isabelle Bloch, Pietro Gori
Computer Vision
Circuit-MLLM: Topological Logic-Guided Latent-Space Visual Reasoning for Circuit Schematic Understanding
Jinyuan Deng, Yuqi Jiang, Wenjing Huang et al. (+3 more)
Through pre-training on extensive text and image
datasets, current multi-modal large
language models (MLLMs) achieve strong performance on general tasks. However,
circuit schematics present a unique challenge for MLLMs due to their dense component layouts and distinct topological logic, demanding fine-grained structural parsing to extract the electrical semantics. To address this, we propose
Circuit-MLLM, a
multimodal reasoning framework that reformulates
circuit topology analysis as a process of device localization, path tracing, and sequential
reasoning within the latent space. We introduce a
circuit knowledge mining mechanism that deeply aligns the model's latent representations with structurally rich features derived from multi-granularity
circuit vision experts, enabling the model to effectively internalize topological semantics. Building upon these internalized semantics, we devise a
topology-guided sequencing strategy that decouples
reasoning from the rigid raster-scan order, enforcing stepwise inference along the
circuit's topological logic in latent space. Across diverse
circuit analysis tasks,
Circuit-MLLM consistently outperforms strong baselines, notably achieving a 25% higher average score than GPT-5.1, which demonstrates the effectiveness of our framework in
circuit schematic
topology analysis. Code is publicly available at https://github.com/IC-Yuan/
Circuit-MLLM.
Read the full abstract ▾
All authors Jinyuan Deng, Yuqi Jiang, Wenjing Huang, Xin Li, Qi Sun, Cheng Zhuo