Materials Science
Three-dimensional atom-by-atom measurement of interface diffusion dynamics
Doojin Park, Hyesung Jo, Yongmin Kwon et al. (+7 more)
Diffusion at interfaces governs how materials and devices form, evolve and function. It is usually measured through smooth concentration profiles and bulk transport coefficients, which average over the discrete atomic rearrangements. At nanoscale interfaces only a few atomic layers wide, however, such coarse-graining obscures the local three-dimensional (3D) atomic transport, and scalar
diffusion coefficients cannot fully describe species-, layer- and facet-resolved redistribution. The essential observable is the discrete 3D redistribution of atoms themselves, beyond an averaged profile. Here we combine micro-electro-mechanical systems (MEMS) pulse-quench heating with atomic electron
tomography (AET) to reconstruct five 3D atomic configurations of the same coherent Pd@Pt core-shell nanoparticle, each corresponding to a well-defined thermal state. Species-conserving one-to-one assignment links neighbouring configurations, and labelled reference simulations correct the displacement bias caused by indistinguishable same-species atoms. The corrected Pd diffusivities follow Arrhenius behaviour, giving an apparent activation energy of 1.10 eV, consistent with migration-limited vacancy-mediated transport. Atomic-layer transition statistics further show that increasing 3D mobility can coexist with declining net chemical transfer and reveal facet-dependent pathways hidden by the scalar diffusivity. Together, these measurements establish a 3D atom-by-atom framework that connects identity-corrected mobility to species-, layer- and facet-resolved transport at buried nanointerfaces.
Read the full abstract ▾
All authors Doojin Park, Hyesung Jo, Yongmin Kwon, Seokjo Hong, Jaewhan Oh, Youngjoo Whang, Eunjik Lee, Gu-Gon Park, Sang Woo Han, Yongsoo Yang
Cryptography and Security
MATE: Policy-Aware Security Auditing for Mobile Agents via Synthesis-Driven Trajectory Learning
Changyue Jiang, Jiayi Wang, Xin Wen et al. (+3 more)
Mobile
agents powered by foundation models now automate complex, multi-step workflows on real devices, but their trajectories can violate app-specific security policies. Existing trajectory-level defenses rely on LLM prompting or rigid rules, and thus fail to support fine-grained, natural-
language policies that generalize across apps and tasks. In this work, we introduce MATE, a lightweight, policy-conditioned auditor that encodes both
agent trajectories and natural-
language security policies to determine whether a trajectory violates a given policy and to explain why. Treating policies as editable text rather than fixed model parameters allows MATE to handle user-defined and evolving requirements without retraining. To construct MATE, we build a knowledge base by extracting app descriptions, workflows, and policies from hundreds of popular mobile apps worldwide, and synthesizing over 140K semantically realistic, policy-conditioned trajectories with a multi-stage pipeline. We further release MATEBench, a trajectory-level
auditing benchmark with two synthetic subsets and one real-world subset of manually collected trajectories. Models trained with our synthesis-driven trajectory learning achieve over 95% accuracy on MATEBench, retain strong performance on external safety
benchmarks, and audit trajectories from Zhipu's AutoGLM and Alibaba's Mobile-
Agent on real devices with over 95% accuracy, outperforming prior methods by over 20%. MATE shows that practical, fine-grained security
auditing for heterogeneous mobile
agents is both feasible and effective.
Read the full abstract ▾
All authors Changyue Jiang, Jiayi Wang, Xin Wen, Jiarun Dai, Geng Hong, Xudong Pan
Networking
ANI-Gamut: Benchmarking Agent Reliability across the Gamut of Agent-Network Interface Abstractions
Lorenzo Bracciale, Pierpaolo Loreti, Andrea Mayer et al. (+2 more)
Large
Language Model (LLM)
agents are increasingly trusted to operate live networks: they read state, change configuration, and verify the result. A first-order question is left implicit: at which level of abstraction should the
agent operate? We make the interface-abstraction level an explicit, controlled experimental variable, organizing
agent-network interfaces into a spectrum from raw CLI (A0) through bounded wrappers (A1) and standardized model-driven configuration (A2) to typed transactional service intent (A3-T), reconciled source-of-truth automation (A3-R), and their combination (A4). We present ANI-Gamut, a reproducible, open-source playground that exposes the same task at several levels on a single, densely populated brownfield substrate, where many coexisting services share resources and collateral damage actually arises. We instantiate and measure four points of the spectrum and describe the others only at the conceptual level, and record three dependent variables as the substrate is stressed by injected faults: task reliability, collateral damage against pre-existing tenants, and operational cost. In a pilot with small run counts (n=20, n=6 and n=4 per level), the interface level moves reliability and cost sharply: on a live change-set task a raw-shell
agent fails on all six seeds while a typed transactional interface succeeds on all six (paired McNemar p=0.031, on six discordant pairs), at roughly an order of magnitude less cost, as the engineering effort migrates from the
agent to a reusable transaction layer. Collateral damage, by contrast, is absent at every level, whether benign or under faults: in a tenant-isolated substrate the
agents fail safe, and the blast radius is held by the substrate's isolation, which moves the safety question from the
agent to the substrate.
Read the full abstract ▾
All authors Lorenzo Bracciale, Pierpaolo Loreti, Andrea Mayer, Stefano Salsano, Wim Henderickx
Fluid Dynamics
Underwater bubble transport on superhydrophobic cylindrical rod
Rajalingam A, Sunghwan Jung and Pallab Sinha Mahapatra
The transport of bubbles along a curved superhydrophobic surface is not governed by buoyancy alone, but also by the interaction between the
geometric confinement, the capillary and contact-line resistance, and the hydrodynamic resistance. We systematically investigated the capillary number ($Ca$) and Bond number ($Bo$) of the transported bubble, as well as the size of the rod relative to the bubble. An analytical force-balance model was developed to predict transport velocity, accounting for buoyancy, hydrodynamic drag, and capillary resistance. The transition from transport to detachment was identified. The experiments were carried out sequentially and supplemented by axisymmetric numerical simulations to investigate wake-induced interactions between bubbles. Bubble transport is strongly influenced by rod inclination and the bubble-to-rod size ratio, demonstrating that
curvature-induced confinement alters the balance between driving and resistive forces. The analytical model captured the experimental trends with deviations of less than 10% in most cases. The bubble detaches from the inclined rod at a critical Bond number that depends on rod diameter, facilitating the development of a transport regime map. The study also found that the motion of successive bubbles is strongly coupled. When a following bubble enters the wake of a preceding bubble, it speeds up and attains a higher capillary number than the leading bubble. Both experiments and numerical simulations consistently reproduce this wake-mediated acceleration, which shows that bubble transport is governed by both the force balance on individual bubbles and the hydrodynamic interactions between adjacent bubbles. These findings form the basis for controlling bubble transport, detachment, and collective motion on superhydrophobic interfaces.
Read the full abstract ▾
Accelerator Physics
Experimental Verification of Circumferential Bunch Length Variation and Head-Tail Exchange Affecting Microwave Instability in a Storage Ring
Jihong Bian, Xiujie Deng, Arne Hoehl et al. (+5 more)
Classical analyses of microwave instability are built upon the longitudinal adiabatic approximation, which assumes that the bunch length remains constant around the storage ring. However, in a storage ring with small global phase slippage, the bunch length can vary around the ring and some particles can experience head-tail exchange due to the partial phase slippage and transverse-longitudinal coupling. Our theoretical study reveals that these effects can be beneficial for suppressing microwave instability. A new microwave instability threshold evaluation method has been correspondingly proposed to account for these effects. Here we present the first experimental evidence supporting our theoretical analysis. The measurements confirm that the microwave instability threshold can be increased by a factor of up to six compared to the classical prediction in our cases. Our results can also provide practical guidance for the design of extremely short bunch storage rings.
Read the full abstract ▾
All authors Jihong Bian, Xiujie Deng, Arne Hoehl, Wenhui Huang, Arnold Kruschinski, Carsten Mai, Markus Ries, Chuanxiang Tang
Human-Computer Interaction
When Disability Disclosure Travels: Memory, Privacy, and Contextual Integrity in Conversational AI
Atieh Taheri, Mahya Tazike, Patrick Carrington et al. (+1 more)
Conversational AI assistants remember what people tell them, and for disabled people, that often includes disability. We interviewed 12 adults with disabilities in the United States who use LLM-based assistants such as ChatGPT, Claude, and Gemini about when, how, and why they disclose disability to these systems and how this compares with disclosing to people. Using contextual integrity as an analytic lens, we found that participants disclosed by need rather than by name, translating disability into task-scoped instructions; that the same disclosure was judged against two recipients, a non-judging interlocutor and a data-holding company, producing opposite norms; and that
memory features relieved the burden of repeated disclosure while letting disability information drift into contexts where it did not belong. Participants did extensive boundary work to restore context and wanted control over scope, provenance, retention, and access rather than per-utterance toggles. We discuss implications for the design of conversational AI assistants.
Read the full abstract ▾
All authors Atieh Taheri, Mahya Tazike, Patrick Carrington, Jeffrey P. Bigham
Software Engineering
From Code to Requirements: Agentic Reverse Engineering of Business Rules at Enterprise Scale
Garima Agrawal, Prasun Das, Priyanka L et al. (+5 more)
Business requirements for enterprise software systems are rarely captured in structured form; the logic resides instead in source code, configuration files, and institutional
memory. When these systems must be migrated, extended, or audited, the absence of formal requirements artifacts forces teams into expensive, knowledge-losing manual reverse engineering. This paper presents an agentic framework that autonomously generates Business Requirements Documents (bRDs) through reverse engineering of undocumented enterprise software. Seven specialized
agents collaborate to discover user journeys, extract business rules, and synthesize a locked bRD from code, test suites, configuration, and available documentation, supported by static analysis tools, guided by embedded expert practitioner cognitive models, and overseen by human reviewers at controlled escalation points. Measured on actual execution traces across a real enterprise deployment, the framework generates comprehensive bRDs in under nine minutes per service, with extracted rules independently corroborated against real production defect records. In a comparable prior migration to the same target architecture, delivery required over two years; on the present programme the framework achieves a cost reduction exceeding 98% against a baseline derived from actual repository metrics using IFPUG complexity models and industry
benchmark labor rates.
Read the full abstract ▾
All authors Garima Agrawal, Prasun Das, Priyanka L, Minisha N, Akshay Sarvade, Hemath Manivanan, Sravani Joshna, Prasad Kalyansundaram
Statistics Applications
Criminal Justice Risk Assessments with Unreliable Class Assignments
Richard Berk
Quantitative risk assessments have been used for decades to help inform criminal justice decisions. A set of predictors and a categorical response variable are used to train a statistical classifier. The trained classifier can be employed to compute risk scores for a new, unlabeled case for which a forecast is needed, one risk score for each response variable class. In the spirit of Bayes classifiers, the class with the largest risk score conventionally becomes the forecasted outcome label for that case. The largest and next largest risk scores can be very similar or very different. However, when they are much the same, the outcome class is essentially being assigned by a coin flip; the classification is unreliable.
Forecasting error becomes more likely, and concerns about
fairness can arise. This paper shows how Mondrian conformal prediction sets can increase classification reliability when the classifier is indecisive and provide valid estimates of forecast uncertainty. Risk forecasts for a large
dataset of offenders on probation are used to illustrate the issues.
Read the full abstract ▾
High Energy Physics (Theory)
Semiclassical reconstruction of the Toda pole lattice via analytic continuation of the Probabilistic Approach
Jacques Distler and Bahaa Elshimy
We develop a probabilistic construction of Toda field theory correlators from the Coulomb gas representation. Reinterpreting $n$-point functions of primaries as $w$-moments of random variables with respect to a Gaussian measure, we derive constraints on the Toda momenta and coupling constant. $L^2$ and $L^p$ convergence conditions for the associated random variables, triviality constraints on the correlators, and local integrability conditions for the cross-moments are derived. These constraints define an extended region of convergence for the Toda correlators and provide a systematic, probabilistic interpretation of the Seiberg bounds. We proceed to analytically continue the moment indices by employing a Mellin-Barnes representation of the random variables. The local convergence boundaries become meromorphic local integrability divisors, and, in the semiclassical $b\to0$ limit, these divisors reconstruct the Liouville charge pole
lattice. For $A_n$ Toda theory, we show that natural $W_n$-closure conditions in this semiclassical limit select rank-one, Virasoro-like submodules inside the full Toda theory. On these slices, the scalar integrability conditions reduce to Liouville-like divisors and recover the corresponding Toda charge pole
lattices.
Read the full abstract ▾
Computer Vision
ZIL: Zero-shot Image-to-LiDAR Registration
Zijun Li, Xiaotian Sun, Xuelun Shen et al. (+6 more)
Image-to-
LiDAR registration estimates the camera pose of an image with respect to a
LiDAR point cloud. It has diverse applications in autonomous driving,
robot navigation etc. However, state-of-the-art (SOTA) methods still 1) mostly assume same-frame inputs, struggling with the image and point cloud from distant frames; 2) rely on domain-specific training, failing to generalize to unseen scenarios. We propose ZIL, the first foundation model for zero-shot non-synchronized image-to-
LiDAR registration. ZIL encodes the input image and point cloud with the
Vision and Point
Transformers. In addition to regressing the relative pose, ZIL also learns to predict 3D coordinates, which substantially improves the pose accuracy without additional annotations. Interestingly, naive mix-data training cannot enable zero-shot generalization, which requires normalization on both camera intrinsics and the
LiDAR vertical-axis origin. Trained on 7 public
datasets with 1.4M
LiDAR frames, ZIL consistently and significantly outperforms previous SOTA with a single model across 5 in-domain and zero-shot
benchmarks, reducing the translation and rotation errors by up to 87% and 76% (shown in Fig. 1). Code and models are available at https://github.com/ZijunLi7/ZIL.
Read the full abstract ▾
All authors Zijun Li, Xiaotian Sun, Xuelun Shen, Yao Dai, Sheng Ao, Yangyang Shi, Jakob Engel, Zhipeng Cai, Cheng Wang
Adaptation and Self-Organizing Systems
Discrete-time Kuramoto model with phase lag: Linear stability analysis and onset of synchronization
Prashant M. Gade and Shamik Gupta
We investigate the discrete-time version of the Kuramoto model with phase lag, which comprises globally-coupled phase oscillators of distributed frequencies that are evolving under a nonlinear map. In the continuum limit of an infinite number of oscillators ($N\to \infty$), we derive the exact Frobenius-Perron equation for the time
evolution of the single-oscillator probability density, and study linear stability of the incoherent state. Instability signals onset of synchronization. The corresponding synchronization threshold is obtained analytically for the case of a Lorentzian distribution of the oscillator frequencies. The threshold differs from that of the continuous-time Kuramoto model, reflecting the fundamentally different stability conditions for discrete-time maps and continuous-time flows. Beyond synchronization threshold, we observe several interesting nonlinear phenomena: Unlike the classical Kuramoto model, the discrete-time version exhibits periodic and chaotic states. Numerical simulations of the finite-$N$ system confirm the analytical prediction for the synchronization threshold, while highlighting breakdown of the celebrated Ott-Antonsen ansatz invoked to conveniently study the continuous-time Kuramoto model in terms of a low-dimensional description.
Read the full abstract ▾
Functional Analysis
Vector-valued partial sums on unbounded Vilenkin systems
Deyu Chen and Guixiang Hong
Let \(\Gm=\prod_{k\ge0}\mathbb Z_{m_k}\) be a Vilenkin group that is not necessarily bounded, i.e., \(\sup_k m_k=\infty\). We prove that, for every UMD
Banach space \(X\) and every \(1<p<\infty\), the Vilenkin partial-sum operators are uniformly bounded on \(L^p(\Gm;X)\), with a bound depending only on \(p\) and the UMD constant of \(X\), and not on \(\mathbf m\). This resolves an open problem arising from the work of Clément et al.~\cite{ClementDePagterSukochevWitvliet2000} and later recorded explicitly in the book of Hytönen et al.~\cite[p.~362]{HNVWI}. The proof reduces the partial-sum estimate, via a Paley conjugation identity and a tangent-sequence decoupling argument, to a decoupling inequality for
Fourier projections on finite cyclic groups, which appears to be new. The same approach also yields \(\mathcal R\)-boundedness for the family of partial-sum operators associated with the finer block decomposition, thereby resolving another related problem communicated to us by Fedor Sukochev.
Read the full abstract ▾
Populations and Evolution
From daylight to darkness: a nonlocal model of circadian activity cycles
Dan Bell, Timothé Boyer and Hao Wang
An animal's ability to perceive its surroundings, as well as the ecological interactions it experiences, can be strongly influenced by the sleep patterns of both itself and surrounding species. Animals with limited sensory capabilities may struggle to navigate their habitat at night, while in a predator-prey system, it may be detrimental for prey to be inactive while predators are awake and hunting. Here, we develop a general mathematical framework to study these interactions using partial differential equations, incorporating activity cycles and daylight-dependent perception into existing models of animal movement. We show that both the dominant sensory mode and the length of daylight are key factors determining the strength of aggregation in social species. Extending this framework to a two-species predator-prey system, we use game-theoretic approaches to show that the
equilibrium sleep strategy of each species is shaped not only by its own sensory capabilities, but also by those of its ecological opponent. Depending on the sensory capabilities of the two species,
equilibrium strategies may consist of a single sleep pattern or a combination of multiple sleep patterns.
Read the full abstract ▾
Artificial Intelligence
Trustworthy Agentic AI: Failure Modes, Mitigation Strategies, and a Lifecycle Framework for Autonomous LLM Systems
Fayeq Jeelani Syed, Rehan Ahmad, Ali Al Bataineh et al. (+1 more)
Agentic AI systems built on large
language models can plan over multiple steps, use external tools, retain information in
memory, and coordinate with other
agents. These capabilities make them more useful than static
language models, but they also introduce new security and operational risks. Untrusted content from websites, emails, documents, and databases can enter the same context as system instructions; persistent
memory can carry compromised information across sessions; and access to external tools can turn an incorrect model response into a consequential real-world action. This article reviews the trustworthiness of agentic AI across five interconnected dimensions: safety and robustness,
alignment and human oversight, transparency and auditability,
privacy and data governance, and regulatory compliance. It organizes key failure modes, including indirect
prompt injection,
backdoor triggers, goal misgeneralization,
memory contamination, and cross-session data leakage, into a unified taxonomy. It also examines major mitigation approaches, such as instruction hierarchies, context isolation, spotlighting, process-based supervision, constrained tool use, and
privacy-preserving
memory, while distinguishing techniques supported by empirical evidence from those that remain largely conceptual. Building on this analysis, we introduce the Trustworthy
Agent Development Lifecycle (TADL), a six-phase framework covering specification, design, training, evaluation, deployment, and monitoring. For each phase, TADL identifies relevant trust activities, expected evidence, and risk-based decision gates. Although TADL has not yet been empirically validated, it provides a structured foundation for developing and evaluating more secure and accountable agentic systems. The article concludes by identifying gaps in current
benchmarks and outlining priorities for future research.
Read the full abstract ▾
All authors Fayeq Jeelani Syed, Rehan Ahmad, Ali Al Bataineh, Aakriti Adhikari
Cryptography and Security
UBA-ORL: Unlearning-Activated Backdoor Attacks on Offline Reinforcement Learning
Fengyi Wang, Cong Li, Lulu Xue et al. (+3 more)
Offline reinforcement learning (offline RL) enables policy learning from pre-collected static
datasets without online exploration, and is increasingly deployed not only in safety-critical domains such as autonomous driving and robotic control but also in data-mining applications such as recommendation and behavior analysis. While compliance-driven data removal enhances
privacy, it also opens a previously unrecognized attack surface. We introduce UBA-ORL (Unlearning-activated
Backdoor Attack on Offline Reinforcement Learning), the first unlearning-activated
backdoor attack for offline RL: in the evaluated settings, the attack is substantially suppressed after normal training and becomes pronounced after a compliance-driven deletion (unlearning) request. UBA-ORL employs a dual-sample mechanism: alongside
backdoor trajectories (BD) that link a trigger to malicious actions under inflated rewards, the attacker injects camouflage trajectories (CM) sharing the same trigger pattern but preserving benign actions with equally high rewards. During training, BD and CM provide competing supervisory signals; upon a legitimate deletion request on the CM subset, the residual BD signal can re-dominate, reactivating the
backdoor on demand. Empirical results show that UBA-ORL achieves controllable activation under the evaluated offline-RL configurations, while no-trigger return changes vary by configuration, exposing a previously overlooked security risk in compliance-driven offline RL platforms. We urge the community to develop joint pre-/post-unlearning
auditing mechanisms for compliant unlearning services.
Read the full abstract ▾
All authors Fengyi Wang, Cong Li, Lulu Xue, Qiyu Leng, Ziqi Zhou, Peijin Guo
Mesoscale and Nanoscale Physics
Electrical Probing of Dark Excitons through Microwave Permittivity
Alex Stram, Zhida Liu, Ziheng Zhang et al. (+17 more)
Excitons in atomically thin
semiconductors are almost always probed through their optical signatures, because the short lifetimes of these transient quasiparticles are generally assumed to preclude electrical detection. Here we show that photoexcited
excitons in monolayer tungsten disulfide produce a large, optically tunable permittivity at gigahertz frequencies, and that the effect provides a contact-free electrical route to imaging dark
excitons at the nanoscale. Using
laser-illuminated microwave impedance microscopy, we find that high-purity encapsulated flakes exhibit a purely dielectric response resonant with the
exciton spectrum, whereas defect-rich samples are governed by conventional photoconductivity. Spatial mapping of
diffusion and sublinear power dependence identify long-lived dark
excitons as the dominant contributors, and first-principles modelling of
exciton polarizability reproduces the measured susceptibility. Our results establish
excitons as optically tunable dielectric elements and introduce microwave microscopy as a direct electrical probe of dark-
exciton transport with sub-100 nm resolution.
Read the full abstract ▾
All authors Alex Stram, Zhida Liu, Ziheng Zhang, Yangchen He, Xuejian Ma, Kyoung Pyo Lee, Lisa Frammolino, Fuxiang Chen, Robert J. Boyd, Sirapas Tangton, Anand Swain, Kan Yao, Kenji Watanabe, Takashi Taniguchi, Yuebing Zheng, Chih-Kang Shih, Daniel Rhodes, Li Yang, Xiaoqin Li, Keji Lai
Signal Processing
TiamiTwin: A Digital Twin for Bistatic ISAC Drone Sensing, Validated Against Measurements
Mehdi Zafari, Saeede Enayati, A. Lee Swindlehurst et al. (+1 more)
Monitoring lower airspace over critical infrastructure using cellular signals of opportunity is highly practical because transmitters are pre-deployed, licensed, and continuously active. Digital twins can evaluate the feasibility of such integrated sensing and communication (
ISAC) architectures, but their predictive accuracy must be validated against real-world data. This paper reports validation results for TiamiTwin, a digital twin developed for bistatic
ISAC drone sensing, using empirical measurements from an operational 5G deployment featuring a commercial band n41 gNB and a receiver separated by 572.8 m over a non-line-of-sight (NLOS) channel. TiamiTwin incorporates three parallel channel representations evaluated on a 240-subcarrier grid: the 3GPP TR 38.901 (Release 19) bistatic
ISAC model, a ray-traced site model, and the captured field measurements. Empirical results demonstrate that both statistical and ray-tracing models under-predict the measured root-mean-square (RMS) delay spread by approximately a factor of three. Furthermore, target reflections sit 68 dB below static clutter in power, making target detection entirely dependent on Doppler separation to isolate the drone from zero-Doppler background returns. Despite this severe clutter environment, the target remains separable along 88% of the flight path in the delay, Doppler, or joint delay-Doppler domains.
Read the full abstract ▾
All authors Mehdi Zafari, Saeede Enayati, A. Lee Swindlehurst, Amitav Mukherjee
Combinatorics
Chromatic symmetric functions for annular webs
Jaeseong Oh and Seokbeom Yoon
We introduce a combinatorial definition of chromatic symmetric functions for annular webs. We prove their
symmetry by constructing a web analogue of the Shareshian–Wachs involution and show that they coincide with the symmetric functions associated to annular webs via Turaev's isomorphism. We then derive explicit formulas for their hook Schur coefficients. We also introduce web LLT functions, whose hook Schur coefficients admit positive Laurent-
polynomial formulas. These formulas yield a combinatorial expression for the coefficients of the HOMFLY–PT
polynomial of an annular web.
Read the full abstract ▾
Human-Computer Interaction
Explanation Navigator: Rectifying Out-of-Scope Human Interpretations of Leaky AI Explanations through Conversational Guidance
Yueqing Xuan, Kacper Sokol and Danula Hettiachchi
As explanations of artificial intelligence systems proliferate, their recipients must grasp not only what they convey but also recognise what they cannot. We conducted an interview study with nine participants to examine how explainees reason when their information needs exceed the scope of available explanations. Participants often unwittingly confabulated explanatory insights when relevant information was missing from the explanations, not recognising the inherent limitations thereof. We characterise such explanations as leaky explanations – simplifications that strive to hide complexity yet whose correct interpretation hinges on understanding of the concealed details. To address out-of-scope interpretations we propose Explanation Navigator: a conversational interaction framework that detects mismatches between users' information needs and explanations' content, elucidating pertinent yet implicit details and providing complementary explanations for unmet information needs. An online study with 316 participants showed that our approach allowed explainees to recognise and rectify confabulated explanatory insights, guiding them towards developing correct understanding.
Read the full abstract ▾
Computer Vision
DOA-SORT: Directional Occlusion-Aware Multi-Object Tracking with Distributional Observations
Hao Wang
Identity association in multi-object tracking (MOT) is vulnerable to partial occlusion, truncated detections, and fluctuating confidence scores. Existing motion-dominant trackers commonly represent occlusion as a scalar penalty. This treatment misses the directional observation bias caused by occlusion: left, right, top, and bottom occlusions distort the location and shape of a detection in different ways. We propose \ours{} (Directional Occlusion-Aware SORT), an online and training-free tracker that models these biases explicitly. First, it infers a soft front–back ordering from box overlap and relative bottom positions, and estimates directional occlusion coverage and depth. It then constructs a mixture of one clean and four directional occlusion observation components. The model uses a five-dimensional observation comprising box center, area, confidence, and aspect ratio, and adapts observation noise to predicted occlusion and detection confidence. The directional mixture likelihood is used in high-confidence association, low-confidence association, and track recovery; ambiguity penalties and local order-consistency swaps further reduce identity errors among nearby objects. On the DanceTrack validation split, \ours{} improves HOTA from 63.00 to 66.34, AssA from 45.10 to 49.57, and IDF1 from 62.19 to 65.28 over OA-SORT with the same detector and evaluation protocol. The gains are concentrated in association quality while detection accuracy remains stable. Additional local evaluations on MOT17 and MOT20 train splits characterize cross-
dataset behavior under the same no-ReID tracking protocol.
Read the full abstract ▾
Computation and Language
Analyzing Public Discourse on Urbanism: Topic Clustering, Sentiment Analysis and Retrieval-Augmented Generation using YouTube Comments
Jakob Morales, Monica Hegde and Fayeq Jeelani Syed
Online discourse about urban issues - walkability, cycling infrastructure, public transit, housing density, and street safety - is voluminous but unstructured, and existing city-evaluation tools capture none of it. We present a pipeline and conversational system that combines geographic entity resolution, topic modeling, sentiment analysis, and
Retrieval-Augmented Generation (RAG) over 22,788 chunks of YouTube transcripts and comments spanning 309 North American cities. Beyond the system itself, our contribution is a set of measurements about what happens when standard NLP components meet short, informal, geographically ambiguous text. A Twitter-tuned RoBERTa classifier outperforms a VADER lexicon baseline by 12.6 macro-F1 points (0.589 vs. 0.464; McNemar p = 0.0001), but both models collapse on the neutral class, which dominates urbanist comment traffic; annotators disagree on the same class (Cohen's kappa = 0.53). Dense
retrieval beats a TF-IDF baseline at every cutoff (P@5 0.790 vs. 0.560), and
video-level relevance proxies understate chunk-level precision by a wide margin (0.660 vs. 0.94 under human rating). For groundedness evaluation, we find BERTScore unusable when a multi-sentence generated summary is compared against a single short comment - scores are nearly flat regardless of relevance - and show that ROUGE-1-based groundedness is a paraphrase-driven lower bound rather than a
hallucination rate. These findings generalize beyond the urbanist domain to any RAG system built over short user-generated documents.
Read the full abstract ▾
Systems and Control
Recursive Parameter Identification of Nonlinear Stochastic State-Space Models via Sequentialized Ensemble Kalman Inversion
Farzaneh Barat, Rachel Carter, Sara Wilson et al. (+1 more)
Recursive parameter identification in nonlinear
stochastic state-space models is challenging because unknown parameters affect the measurements through latent-state dynamics. This paper develops a sequentialized ensemble Kalman inversion method for recursive parameter identification from streaming measurements. The proposed method represents the parameter posterior by an evolving ensemble and updates it sequentially as new measurements become available. For each parameter ensemble member, implicit particle filtering is used to approximate the predictive observation statistics required for parameter correction. This formulation enables recursive parameter learning. The method is evaluated on a strongly nonlinear
benchmark, with comparison to several existing methods, and then on a nonlinear double-capacitor lithium-ion battery model. The numerical results demonstrate accurate recursive parameter identification and latent-state estimation.
Read the full abstract ▾
All authors Farzaneh Barat, Rachel Carter, Sara Wilson, Huazhen Fang
Systems and Control
An Adaptive-Sampling Control Framework for Constrained Linear Systems with Robust Safety Guarantees
Spencer Schutz, Charlott Vallon and Francesco Borrelli
Adaptive-sampling control balances control performance with resource efficiency. However, existing methods either fail to guarantee robust constraint satisfaction during rate transitions or require computationally expensive online optimization. This paper proposes an adaptive-sampling control framework for linear systems subject to polytopic state and input constraints and bounded additive disturbances. Given a time-varying reference control update rate provided by a reasoner, our framework continuously calculates Model Predictive Control (MPC) update rates that ensure robust constraint satisfaction at all time steps. Offline, robust M-step hold control invariance is used to precompute invariant sets for a list of update rates and transition sets between them. Online, these sets are used in real time to guarantee recursive feasibility and finite-time transitions to the reference update rate. The utility of the architecture is demonstrated in a cruise control simulation.
Read the full abstract ▾
Artificial Intelligence
Hapi: A Multivariable Land-Surface Transformer for Medium-Range Hydrological Forecasting at Continental Scale
Hong Zhang, John K. Hutchison, Rao Kotamarthi et al. (+7 more)
Accurate flood forecasts several days in advance are essential for flood control, water-resource management, and emergency response. A central challenge is to produce high-resolution forecasts across continental domains where hydrological behavior varies widely from place to place. We developed Hapi, a U-Net Swin
Transformer that uses fine three-dimensional patches and hierarchical shifted-window
attention to forecast river discharge, surface runoff, snow water equivalent, and soil wetness index across the contiguous United States. The model produces medium-range forecasts (24–72~h) at $0.05^{\circ}$ resolution with adaptive task weighting and required only 0.11 seconds for a four-variable 72-h CONUS forecast on one A100 GPU. In a held-out 2024 potential-skill evaluation with ERA5-Land inputs prescribed over the forecast horizon, Hapi achieved the highest F1-score for floods in 20 of 21 comparisons across seven GloFAS return periods and three forecast leads. Independent validation against observed daily discharge at 3{,}881 U.S. Geological Survey gauges showed that Hapi achieved the highest median Nash–Sutcliffe efficiency at every lead, supported by regional-cluster bootstrap intervals. In a matched 24-h comparison of loss formulations, adaptive task balancing produced the lowest discharge errors and the highest F1-score for floods.
Read the full abstract ▾
All authors Hong Zhang, John K. Hutchison, Rao Kotamarthi, Jeremy Feinstein, Haiwen Guan, Romit Maulik, Ross M. Alexander, Vijay P. Ramalingam, Jason Stock, Thomas Wall
Machine Learning
Autonomous Model Lifecycle Management for Digital Twin-Based Manufacturing Control
Zhengyang Gu, Thomas Cook, Fredaljohn Rohrbaugh et al. (+2 more)
Manufacturing AI systems must autonomously adapt to continuous distributional shift from raw-material variability, ambient changes, and equipment aging, under strict safeguard and operator-trust requirements where model failures risk physical damage. This paper presents a closed-loop Cyber-Physical System (CPS) for autonomous model lifecycle management in automotive manufacturing, deployed since 2023. The system manages product-specialized model pairs: a sequence-to-sequence physics model (LPP) serving as a digital twin, and a deep Reinforcement Learning (RL) control policy (LCP) trained against it. Per retraining cycle, multiple model variants spanning architecture families and RL
algorithms compete; only the best-scoring candidate advances. A Conductor orchestrator autonomously manages plant-wide model inventories with dependency-aware retraining and Proportional-Integral-Derivative (PID) fallback. Reflecting the principle of Human-Centric Intelligence, the LCP composite score embeds an operator-trust gate penalizing policies deviating from established practice; without it, 23% of policies are rejected by operators despite passing accuracy thresholds. Across multiple facilities, LCP-controlled processes achieve process stability improvements of 28-45% over uncontrolled baselines with zero safety incidents.
Read the full abstract ▾
All authors Zhengyang Gu, Thomas Cook, Fredaljohn Rohrbaugh, Joseph E. Hernandez, Chris Couch
Computation and Language
LLaDA-PRM: A Bidirectional Step-Level Reasoning Evaluator
Yiming Feng, Naihao Deng, Yulong Chen et al. (+1 more)
Step-level
reasoning evaluators are commonly based on autoregressive
language models, whose causal
attention restricts each step representation to the problem, previous steps, and the current step. Yet, when the complete solution is available, the validity of an earlier step may become clearer only through its downstream consequences. We validate this hypothesis through a controlled 54-run comparison of causal and bidirectional LLaDA evaluators at 1B–3B scale, changing only the self-
attention mask, and find bidirectional
attention yields consistent improvements. Building on this finding, we introduce \prm{}, an 8B bidirectional evaluator that reaches 88.8 step-level F1 on MR-MATH-invalid and 83.8 on the out-of-distribution MR-GSM8K original-question subset, outperforming ReasonEval-Llemma-34B by 11.3 and 10.3 F1 points, respectively. \prm{} also remains effective when evaluating incomplete
reasoning traces in online settings, outperforming the strongest baselines on both
benchmarks by a large margin. We further show that \prm{} provides an effective training-data selection signal, improving Mistral-7B performance on MATH-500.
Read the full abstract ▾
All authors Yiming Feng, Naihao Deng, Yulong Chen, Rada Mihalcea
Complex Variables
A Product Principle for Harmonic Schwarz Lemmas: Boxes, Polydiscs, and Metric Geometry
Miljan Knežević and Miodrag Mateljević
We establish an exact product principle for the Euclidean operator norm of differentials of harmonic maps. For a bounded domain \(G\subset\R^m\) and \(p\in G\), let \(M_G(p)\) denote the supremum of \(\|dF_0\|\) over harmonic maps \(F:\D\to G\) with \(F(0)=p\). For bounded domains \(G_j\subset\R^{m_j}\), we prove \[ M_{G_1\times\cdots\times G_N}(p_1,\ldots,p_N)^2 =\sum_{j=1}^N M_{G_j}(p_j)^2. \] The
theorem separates the geometry of the individual factors from the Euclidean geometry of the product: the factor extremal constants combine by a sum-of-squares law, while equality is governed by a single compatibility condition, namely a common maximizing direction for the component differentials. Neither convexity nor attainment of the factor suprema is required. Combining the product principle with the sharp interval and disk factor problems yields exact operator-norm estimates and all equality cases for harmonic maps into boxes and polydiscs. In both families, for every extremal map, the real differential at the origin has one-dimensional image. The same factor constants also define coordinatewise metrics for which the harmonic contraction estimate is sharp when the source disk is equipped with its Poincaré metric of
curvature \(-1\). For boxes, the resulting metric is complete and equals twice the restriction of the Kobayashi-Royden metric of the product of vertical strips. For polydiscs, the harmonic product metric is pointwise maximal among contracting metrics of the form \(\max_j a_j(p)|v_j|\), with \(a_j(p)>0\). It is strictly smaller than twice the Kobayashi-Royden metric on every nonzero tangent vector, and its induced path metric is incomplete.
Read the full abstract ▾
Lattice Field Theory
First-Principles Determination of the QCD Contribution to the Axion-Photon Coupling Using Domain-Wall Fermions
Tian Lin, Hao-Yang Cheng, Xu Feng et al. (+3 more)
The
axion-
photon coupling, crucial for experimental
axion searches and tests of the strong CP solution, receives a substantial model-independent contribution $\mathcal C_{\rm QCD}$ from QCD dynamics. Next-to-leading-order
chiral perturbation theory (NLO ChPT) in different frameworks has yielded puzzling discrepancies of up to $8\%$, motivating precise first-principles calculations. We present an independent
lattice QCD determination using a method complementary to the recent background-field calculation. Computing pseudoscalar-to-two-
photon three-point functions and exploiting anomalous Ward identities, we separate $\mathcal C_{\rm QCD}$ into an exact anomaly contribution and a light-
quark-mass-suppressed correction. The latter is computed using domain-wall
fermions, whose excellent
chiral symmetry strongly suppresses discretization effects. Working on two near-physical $N_f=2+1$ ensembles with continuum extrapolation, we obtain $\mathcal C_{\rm QCD}^{\rm IS}=1.619(30)$ (isospin-symmetric), $\mathcal C_{\rm QCD}^{\rm IB}=0.347(22)$ (isospin-breaking), and $\mathcal C_{\rm QCD}=1.965(35)$ in total. While direct comparisons with published NLO ChPT predictions reveal apparent tensions, we identify their sources and show that the ChPT results can be reconciled with our
lattice determination. Our result provides a first-principles
benchmark for the QCD contribution to the
axion-
photon coupling and a quantitative test of competing ChPT descriptions.
Read the full abstract ▾
All authors Tian Lin, Hao-Yang Cheng, Xu Feng, Lu-Chang Jin, Chuan Liu, Qi-Yuan Luo
Computation and Language
COT-TTS: Audio Context-Aware Text-to-Speech with Chain-of-Thought Reasoning
Weizhen Bian, Sitong Cheng, Rongxiu Zhong et al. (+9 more)
Recently, text-to-
speech systems have made significant progress in
speech expressiveness and controllability. However, the speaking style of generated
speech typically relies on clear user-specified instructions. In natural conversations, speaking style should be naturally inferred from the preceding conversational context. Therefore, we propose COT-TTS, a context-aware,
reasoning-based text-to-
speech task. Given historical conversation audio, target text, and a reference
speech, the system should comprehend the conversational context, infer an explicit intermediate
reasoning, and finally synthesize the target
speech with the specified timbre. To support this task, we constructed a large-scale bilingual conversational
speech dataset comprising 9 million training samples, including a high-quality subset of 1 million samples. We further constructed a source-disjoint
benchmark with 800 human-verified samples and established strong task-specific baselines. Additionally, we developed end-to-end autoregressive models with parameter sizes of 0.6B and 1.7B, generating
emotion-labeled transcripts, editable
speech style inferences, and
speech tokens. Experimental results show that the proposed model achieves performance comparable to large-scale baseline systems with significantly fewer parameters. At the same time, the model performs well in terms of duration consistency and emotional consistency, and can generate appropriate emotional, stress, and rhythmic variations based on the conversational context. To facilitate future research, we will publicly release the data construction pipeline,
dataset, trained models, and related resources. The demo page and additional resources are available at https://luckybian.github.io/COT-TTS
Read the full abstract ▾
All authors Weizhen Bian, Sitong Cheng, Rongxiu Zhong, Jiahao Pan, Liumeng Xue, Boyi Kang, Shilei Zhang, Jinglei Liu, Yue Wang, Junlan Feng, Bei Liu, Wei Xue
Artificial Intelligence
Building Trustworthy Mental Health Benchmarks on Bluesky: A Validation-Aware Weak-Supervision Framework
Gaurab Chhetri, Anandi Dutta and Subasish Das
Decentralized social media platforms create new opportunities and challenges for computational mental health research because data access, moderation, labeling, and deployment responsibilities are distributed across multiple technical and governance layers. This paper presents a validation-aware weak-supervision system for constructing and evaluating suicidal ideation (SI) and broader mental health (MH) disclosure
benchmarks on Bluesky, a decentralized social media platform built on the AT Protocol. The system integrates public firehose collection, task-specific lexicon filtering, Llama-3-8B-assisted binary annotation, human-adjudicated validation subsets, and
transformer-based model benchmarking. Using this pipeline, we construct two task-specific corpora containing 8,346 SI-labeled posts and 9,988 MH-labeled posts. The evaluation shows that model performance depends strongly on both task definition and validation protocol. BERT+LSTM achieves the highest SI stratified cross-validation F1-score, RoBERTa achieves the strongest SI holdout F1-score, and DistilRoBERTa achieves the best MH cross-validation F1-score. Human validation reveals different weak-label failure modes across tasks, with SI labels dominated by false negatives and MH labels dominated by false positives. These findings show that decentralized social media can support reproducible mental health benchmarking, but only when system design, label provenance, validation strategy, and deployment constraints are evaluated together.
Read the full abstract ▾