Science & Mathematics

Unmanned aerial vehicle (UAV) platforms with integrated sensing and communication (ISAC) are prime candidates for surveillance missions, in which a dual-function drone tracks a mobile target and alerts a ground station when the target enters a protected region. Such alerts generate event-triggered communication traffic. By observing the change in the traffic, an eavesdropper (Eve) can therefore learn in real time whether the target is detected inside the protected region without decoding any message. Existing physical-layer security techniques protect the message content or hide that a transmission occurs, but the composition of the traffic remains exposed. We formulate the threat as traffic-composition covertness and propose a keyless diffusion-based transmission scheme that hides the triggered traffic. The scheme superposes the beamformed information streams of the UAV with a message-agnostic safeguard waveform generated by a conditional denoising diffusion implicit model. Training shapes the safeguard distribution such that triggered and non-triggered slots become statistically indistinguishable at Eve. The legitimate user shares no secret key with the UAV. We formulate the joint design as a rate maximization under covertness, reliability, and sensing constraints, with the sensing accuracy measured by the posterior Cramér-Rao bound (PCRB) of the beam-dependent tracking problem. Our numerical results show that, with multiple transmit antennas and an Eve signal-to-noise ratio (SNR) of $5$~dB, a learned Eve stays near the blind guess of $0.5$ for every safeguard beam split, a detector that knows the safeguard model reaches $0.71$ to $0.73$ inside Bob's main lobe and $0.53$ to $0.55$ outside it, while the average position PCRB varies only between $0.59$ and $0.70$~m$^2$.

Recommending bacterial cultivation conditions from sequence data is a challenging structured problem characterized by missing standardized benchmarks and limited use of sequence structure. To address this, we introduce DSNI, a benchmark of 65,023 bacterial strains with phylogeny-aware splits, pairing 16S rRNA sequences with four cultivation conditions (media, temperature, pH, oxygen). Exploiting the mosaic structure of 16S rRNA, where hypervariable regions support discrimination and conserved regions support primer design, we propose a Variability-Gated Encoder that learns position-dependent importance weights conditioned on biological features (region type, position, and local complexity). The gating module adds minimal overhead (1,280 parameters,

Lateral fluorination has emerged as a powerful design variable in the design of polar liquid crystals, but its interplay with terminal-group identity remains poorly understood. Here we report a 40-member matrix of biphenyl-ester-phenyl mesogens that independently varies lateral core fluorination and terminal-group identity. Increasing lateral fluorination generally suppresses apolar mesophases and clearing points, however this same change can enable polar nematic phase behaviour in conjunction with suitable terminal units. Graph Fourier analysis of conformer averaged atomic charges shows polar materials occupy a high-frequency region of charge-spectral space, consistent with a bond-scale electrostatic motif. Different terminal units can bias this towards higher or lower frequencies, and in doing so can enable polar order to emerge.

Natural-language instructions often rely on high-level abstractions, such as asking an agent to ``stack blocks in a pyramid.'' Interpreting abstract instructions requires models to both identify the abstraction intended by the speaker, and unroll it into the many concrete actions that instantiate it. Language models struggle here: prior work has shown that performance degrades as the level of abstraction in the instruction rises. To bridge the gap between natural language abstraction and concrete actions, we propose to borrow from coding environments the notion of a {\em domain-specific language} (DSL), an explicit representation layer whose documented primitives name recurring abstractions that expand into concrete actions. We demonstrate the efficacy of DSLs for NL interpretation on \textsc{Hexagons}, a visually-grounded instruction-following task graded for abstraction. We further ask whether language models can induce such DSLs themselves from domain data. Our results show that while code generation consistently improves over direct inference, code generation over a human-curated DSL yields dramatic gains, up to 10 absolute F1 points, reaching the human skyline, with gains concentrated at {\em higher} abstraction levels. We confirm this pattern on the \textsc{Text2CAD} benchmark. At the same time, model-induced DSLs recover only part of this gain, remaining inconsistent across runs and short of their human-curated counterpart. Overall, the right representation closes much of the gap caused by high abstraction --- but inducing that layer of representation automatically is where models still fall short, and where we see a strong opportunity for the community to make progress.

Geospatial foundation models (GFMs) differ widely in pretraining objective, modality and resolution, and none leads across benchmarks. We ask whether their representations are complementary, and whether a single encoder can inherit that complementarity. We introduce Pangaea-Latents, an extension of the Pangaea benchmark providing frozen embeddings of $5$ GFMs over its $11$ datasets, along with reproduced baselines, and find GFM embeddings to be neither redundant nor disjoint. We then propose GeoMixture, a preliminary framework to align a ViT to several GFMs per dataset through a multi-representation alignment (mREPA) objective. Our encoder ties the best GFM in average rank across nine Pangaea datasets, trading peak performance for cross-task consistency.

Prior-data fitted networks (PFNs) have recently gained popularity as an end-to-end approach to amortized Bayesian inference. Yet, their statistical behavior for semi-parametric targets, such as means under missingness or causal effects, remains poorly understood. In this work, we develop theoretical foundations for debiased PFNs to perform inference on finite-dimensional targets in general semi-parametric settings. First, we show that, despite being Bayes-optimal with respect to the target, an idealized PFN can exhibit a posterior concentration rate that implicitly depends on the concentration rates of (latent) nuisance functions and can become slower as the prior becomes more diverse. Then, to speed up the target posterior concentration rates, we introduce two generic debiasing strategies based on efficient influence functions: (i) a label correction and (ii) a prior correction. Finally, we show that the debiased PFNs (i) & (ii) possess concentration rates that asymptotically match a convergence rate given by traditional frequentist bias-corrected estimators. In this way, we establish a novel connection between amortized Bayesian inference with PFNs and frequentist semi-parametric efficiency theory. Under additional assumptions, this allows us to guarantee local and global optimality of PFN-based estimators (namely, the Bernstein-von Mises property and minimax optimality, respectively). Therefore, our work provides guidelines for designing debiased PFNs for semi-parametric inference with principled and provable asymptotic guarantees.

Large language model (LLM) agents increasingly use persistent memory to personalize responses, and personal attributes in this memory pose a privacy risk if exposed. We ask how consulting memory during dialogue changes what users go on to say, and thus what the memory comes to reveal. In multi-session simulations, a simulated user with 24 personal attributes talks with a support agent that does or does not consult memory, and in both settings attributes are inferred from the full accumulated record. Across four support models, memory raises response quality but also empirical leakage, the share of attribute values correctly inferred from the record, by a margin that widens over sessions. The cause is a loop in ordinary, non-adversarial assistance. Memory-conditioned responses carry earlier values forward, often as presuppositions, and users affirm them or leave them uncorrected, so that cues and the agent's own guesses become recorded values. The loop reinforces itself, and removing the agent's mentions of an attribute lowers its later disclosure about twice as much as single-turn effects predict. Users cannot leave it on their own, since correcting a value discloses the attribute, declining to discuss it shifts later predictions toward sensitive values, and incorrect values left in place become entrenched. The loop is best weakened at the agent's response. Instructing the agent not to state or presuppose attributes brings leakage to about the no-memory level while keeping most of the quality gains, whereas filtering or anonymizing what users say or what is stored gives up much of them.

Instruction-based image editing models can now add, remove a specified number of objects or change their category while preserving the scene. Yet these models often fail to produce the correct number, and these failures grow sharply as scenes become more complex and the requested count increases. Current benchmarks largely miss this problem: they score edits with holistic similarity metrics or with vision-language model (VLM) judges that are themselves poor at counting, so numerical errors go unnoticed or are graded incorrectly. As a result, the true numerical ability of today's editing models has never been accurately measured. We address this issue with ABACUS, a new benchmark, evaluation protocol and dedicated scoring metric for numerical image editing. Our protocol is judge-free: no VLM or MLLM scores an edit or counts objects (a VLM only confirms the category of rare ambiguous crops). Instead, it verifies each edit at the instance level against panoptic ground truth confirmed by detector consensus, checking both that the exact requested count was achieved and that the rest of the scene was preserved, across the add, remove and change operations. By evaluating leading image editors, our method reveals substantial numerical editing failures that worsen with increasing scene complexity and object count; failures that prior benchmarks often overlooked or misjudged.

Generating graph signals is an important task in machine learning, and incorporating the underlying graph into a generative model enables structurally-informed distribution modeling. However, said graph is often noisy or imperfectly estimated, and structural errors can affect the generative process and the resulting signal distribution. While recent work has established stability guarantees for graph neural ordinary differential equations, the robustness of graph-aware reference dynamics, as used in topological Schrödinger bridge models, remains largely unexplored. Here, we investigate the structural stability of continuous-time generative models whose dynamics combine a graph filter with a learned graph neural network. We first derive a Wasserstein stability bound that quantifies the effect of graph perturbations on the generated signal distributions. Informed by this analysis, we then propose a principled framework for designing graph filters that balance structural stability with induced signal smoothing. We characterize the optimal filter design and provide an explicit family of polynomial filters with provable stability guarantees. Experiments on synthetic graphs and real-world fMRI signals support our theoretical findings, illustrating the enhanced structural stability of the proposed filters while preserving generative performance.

Physics-informed neural networks (PINNs) supervise on the residual of a known differential operator. Many high-stakes inference problems admit no such operator: clinical signals, behavioral measurements, materials properties. We introduce the Manifold-Informed Neural Network (MINN), a PINN-parallel paradigm that replaces the physics operator residual with a manifold operator residual: a discrete Poisson equation $\mathbf{L}_G \mathbf{H} = s_φ(\mathbf{x}, \mathbf{m})$ on a $k$-nearest-neighbor graph in the learned latent space, with a learned, task-aware source. The construction is a structural parallel to PINNs (squared operator residual plus label data, in place of operator residual plus boundary data); the $s{=}0$ limit recovers the homogeneous Laplace equation $\mathbf{L}_G\mathbf{H}=\mathbf{0}$ (classical manifold regularization solves the same equation but minimizes the Dirichlet energy $\operatorname{tr}(\mathbf{H}^\top\mathbf{L}_G\mathbf{H})$, a different penalty form). The source $s_φ$ takes input features and auxiliary metadata directly, breaking the homogeneous smoothness prior that limits classical Laplacian penalties; gated Feature-wise Linear Modulation separately fuses metadata into the prediction branch. We evaluate MINN on three-class classification of potassium, calcium, and magnesium from 12-lead ECGs in MIMIC-IV-ECG, against twelve baselines spanning supervised and tabular methods, ECG foundation models, and manifold/graph approaches. MINN attains the highest K$^+$ Macro-F1 of 0.61 (best baseline 0.55), with consistent gains on Ca$^{2+}$ and Mg$^{2+}$ and a 31% relative improvement at 1% labeled data. Six targeted ablations show that the inhomogeneous Poisson formulation, not graph smoothing alone, drives the gain.

Streaming video understanding requires a model to decide not only what to answer but also when to answer. Evidence arrives progressively, and recent benchmarks identify premature responses, issued before the queried event has concluded, as a recurring failure mode. Existing approaches often learn response timing through additional training, while training-free alternatives typically rely on scalar gating signals that do not explicitly represent when an answer becomes final. We organize chunk-wise streaming errors into three patterns: sampling variation can make an answer unstable, a stable answer may lack support in the observed video, or an answer may be supported by the current evidence while the event remains unfinished. We propose PACE (Plan, Attend, Check, and Emit), a training-free framework designed around these three failure modes. Given a question, PACE drafts a plan specifying what to observe and what will signal completion. As the stream unfolds, it invokes an off-the-shelf video LLM at informative moments, evaluates each draft for stability, grounding in its cited frames, and completion, and records verified observations in an event ledger. Experiments on ProactiveVideoQA and OVO-Bench Forward show that PACE improves over the official protocols by up to 5.9 average PAUC points and 6.2 average accuracy points across proprietary and open-source backbones, yielding state-of-the-art performance.

Deterministic industrial networks often present an operational trade-off between maintaining microsecond timing of primary control loops and capturing security telemetry. This work investigates the engineering feasibility of the Operational Technology Supervisor (OTSV), a three-tier passive monitoring framework designed to resolve this visibility challenge within timing-sensitive grid-edge infrastructure. By utilizing a Layer-3-unconfigured passive ingestion mode on bare-metal hardware via a physical switched port analyzer (SPAN) loop, the deep packet inspection engine is isolated from the production network, implementing a software-enforced unidirectional export policy. Performance validation conducted over a continuous audit comprising 32,927 active-monitoring round-trip time (RTT) samples yielded stable latency distributions across a laboratory power-monitoring loop running Modbus TCP and IEC 61850 MMS protocols. The measured latency delta between the unmonitored baseline and the active monitoring state was less than 0.12 milliseconds, supporting the feasibility of continuous high-fidelity security observation alongside stringent grid-edge metering timing bounds without measurably degrading the monitored reporting path.

Factual sycophancy, a language model's deference to a user's false beliefs, is usually understood as a bid for the user's approval at the expense of truthfulness. We argue that this framing over-emphasizes the user and propose viewing sycophancy as a failure within the broader and generally useful process of in-context updating, by which models revise their answers based on information supplied during an interaction. We support this view with behavioral and mechanistic evidence. First, models' responses to user claims sit within a broader pattern of source-sensitive updating. They revise their answers in response to information attributed to external sources as well, more so when it appears better supported, and their deference to the user persists even when no user approval is at stake. Second, the internal components that support deference to the user are also causally important for in-context updating more broadly. Our results suggest that factual sycophancy arises when in-context updating gives a user's assertion excessive weight relative to its evidence. Because sycophancy is embedded in this process, reducing it risks reducing useful updating as well, a trade-off that calls for evaluations measuring both behaviors and mitigations that decouple them.

We present the Observation Assimilation Transformer (OAT), a geometry-aware neural process that corrects and refines a prior numerical weather prediction (NWP) field into a high-resolution posterior field by assimilating sparse observations. OAT is an encoder-decoder transformer that considers a set of observed locations and their NWP prior to condition predictions at a set of arbitrary unobserved query locations. Its sparse attention mechanism makes it scalable to thousands of station observations and depends on relative positions between locations, never on the absolute coordinates. The model allows for partially missing observations, predictions at arbitrary continuous locations without grid projection, and outputs are parametric distributions. A station-randomized training algorithm is developed with continuous ranked probability score (CRPS) loss. We test OAT on two datasets - one at regional scale with hundreds of stations, and one at continental scale with thousands of stations. On these datasets OAT outperforms several baselines and shows well calibrated predictions. In an illustrative case study, OAT's downscaling capabilities are demonstrated by downscaling ERA5 (~31km) to COSMO (1km), where OAT achieves lower mean absolute error (MAE) and CRPS for temperature and precipitation on test stations compared to COSMO.

Trajectory tracking for compliant robots is challenging due to nonlinear dynamics and modeling uncertainties, often requiring learning-based methods to avoid explicit parameter identification. Iterative Learning Control (ILC) ensures theoretically high precision and preserves the passive compliance of the robot. However, traditional ILC requires a separate learning phase for each new trajectory. Conversely, Reinforcement Learning (RL) generalizes to new trajectories, but practically suffers from high sample complexity and limited tracking precision. This work proposes Reinforced Iterative Learning Control (RILC) that combines the strengths of both approaches to achieve fast adaptation and guaranteed tracking convergence across diverse trajectories. By establishing a concurrent coupling between the two controllers, RILC overcomes the high initial errors and slow adaptation of traditional ILC, enabling high-performance tracking from the very first trials. In this architecture, ILC provides high-quality samples, while RL contributes as a generalizing best-effort policy across iterations. Additionally, a theoretical proof is presented to show that the RL component preserves ILC stability and, under specific conditions, can also reduce the error bound asymptotically. The efficacy of this framework is demonstrated through simulation and experiments on an underactuated 4-degree-of-freedom serial elastic robot using different trajectories and modeling mismatches.