Papers

Behavioral foundation models (BFMs) have recently shown that a single humanoid policy can support diverse whole-body control, but extending such generality to physical interaction remains challenging. We introduce I-BFM, to our knowledge the first BFM for humanoid-object interaction. Rather than relying on task-specific policies or reference tracking, I-BFM learns a shared representation of the coupled dynamics among the humanoid, objects, and their contacts using forward-backward representations and unsupervised reinforcement learning. Given a downstream task reward, the same policy can be directly conditioned on a latent command to execute closed-loop interaction without task-specific policy optimization. To improve interaction control over different time scales, we further train the policy with both short-horizon interaction targets and longer-horizon goal targets. A single I-BFM policy performs carrying, pushing, and kicking, while also supporting goal reaching, motion tracking, stylistic control, and long-horizon task chaining. More importantly, it remains effective after large deviations from nominal execution: on Carry, I-BFM achieves 94.3% nominal success and retains 89.3% success after robot falls, compared with 1.3% for a planning-based baseline. Real-world experiments on a Unitree G1 further demonstrate diverse loco-manipulation behaviors, rapid recovery from interaction failures and external disturbances, and task chaining without task-specific retraining.

Predictive maintenance requires reliable remaining useful life (RUL) estimation. Existing methods mainly follow two paradigms: wear-based aging models that capture cumulative degradation and sensor-driven data models that reflect instantaneous health conditions, each providing only partial information. In this work, we propose a probabilistic fusion framework that integrates wear-based and sensor-based prognostic components through failure probability distributions. Based on explicit structural assumptions linking wear, latent health, sensor observations, and failure, we derive a principled combination rule that enables uncertainty-aware integration with adaptive weighting of the components. Experimentally, we assess this combination rule by learning the wear-based component using a parametric survival model and the sensor-based component using a 1D convolutional neural network (1D-CNN) with a post-hoc uncertainty model. Evaluation on multiple N-CMAPSS datasets demonstrates that the fused model improves point accuracy, preserves the C-index, and produces narrower yet well-calibrated prediction intervals compared to either component alone. The results highlight the complementary roles of wear-based survival model and sensor-based deep learning model, and show that their probabilistic integration provides a structured pathway toward more robust and consistent prognostics over the life-time.

The geometric measure of entanglement of mixed states is investigated based on its relation to spin correlations. We derive an analytical expression for the entanglement of rank-2 mixed states composed of arbitrary multiqubit pure states within a two-dimensional Hilbert subspace. The dependence of the entanglement of rank-2 mixed multiqubit states on the parameters characterizing arbitrary pure states in the mixture is studied both analytically and using quantum computing. For a particular class of mixed states, namely, equal-probability mixtures of an arbitrary multiqubit state and the Schrödinger cat state, we construct quantum protocols for detecting the geometric measure of entanglement based on its relation to spin correlations. We quantify the entanglement using both the AerSimulator and the ibm_marrakesh quantum computer. The effects of quantum-computing errors are also analyzed. The simulation results reproduce the analytical dependencies, while measurements on the quantum device demonstrate good agreement with theoretical predictions within estimated error bounds.

Dense cores and clumps in massive star-forming regions are studied through (sub-)millimetre continuum observations, but inferred masses and temperatures can be biased by limited resolution, interferometric filtering, projection, and uncertain dust temperatures. We quantify how reliably core masses and temperatures can be inferred from ALMA-like continuum data using synthetic observations of such clumps. We post-processed radiation-hydrodynamic simulations of collapsing 1000 M$_\odot$ clumps with RADMC-3D and CASA to generate synthetic ALMAGAL-like images. Cores were defined around sink particles, enabling comparison of masses from the 3D simulations, idealised RADMC-3D maps, and synthetic ALMA images. We quantified how resolution, projection, and viewing angle affect core masses, and tested three dust-temperature estimators (clump-scale L/M, a L/M+flux model, and a supervised machine-learning estimate). Projection effects cause line-of-sight integrated core masses within 1500 au to exceed spherical masses by factors of two to three. The empirical L/M temperature prescription follows the global thermal evolution but misses the core-to-core scatter, producing core-mass errors of a factor of a few. Including local flux improves temperature and mass estimates, while a gradient-boosting machine-learning model suggests further improvement. Synthetic ALMAGAL-like images recover the global core structure, but flux-reconstruction uncertainties introduce mass errors up to a factor of three and limit reliable core-mass recovery to $\gtrsim 0.3$ M$_\odot$, comparable to the ALMAGAL completeness limit. Individual core masses are typically uncertain by a factor of three to five, mainly due to dust-temperature assumptions and flux reconstruction. Reducing these biases requires temperature estimates that include local core information rather than a single clump-scale L/M-based prescription.

We consider the problem of allocating $m$ indivisible items to $n$ agents. For $n \ge 2$, we define $\mu(n)$ as the largest $m$ for which every allocation instance with $n$ agents and $m$ items has an allocation that gives each agent at least her maximin share (MMS). We prove that for general monotone valuations, $\mu(n) = n + \Theta(\log n)$. This holds both for goods and for chores. For arbitrary valuations (not monotone), we prove that $\mu(n) = \log n + O(1)$.

In this paper, we give the complete characterization of the boundedness of the discrete fractional maximal operator $M_\gamma $, defined on $\mathbb{Z}^n$, $n\in\mathbb{N}$, $\gamma\in [0,n)$, between the classical Lorentz sequence spaces $\lambda_p(v)$ and $\lambda_q(w)$, where $0<p,q<\infty$, $\{v(m)\}$ and $\{w(m)\}$ are non-negative sequences defined on $\mathbb{N}$. We first obtain a sharp upper estimate for the nonincreasing rearrangement of the discrete fractional maximal operator $M_\gamma x$, with a supremal operator involving a discrete Hardy operator and the nonincreasing rearrangement of the sequence $\{x(k)\}$, and we show that this estimate is sharp by showing a lower estimate for special sequences. Using this result, we can reduce the characterization of the boundedness of the discrete fractional maximal operator $ M_\gamma$ between the discrete weighted Lorentz sequence spaces to the boundedness of the discrete supremum operator between weighted Lebesgue sequence spaces, restricted to nonincreasing sequences. This problem was investigated in the authors' recent paper, which enables us to obtain the final characterization.

In this study, we focus on large-scale ad hoc networks characterized by a high node density. To achieve energy efficiency, the devices must minimize their data transmission, thereby reducing the total number of bits sent and the associated energy consumption. We consider using ID codes for the address field in the MAC layer and reduce the header size. The work explores a range of use cases and challenges that demonstrate the benefits of incorporating these ID codes into packet routing mechanisms for ad hoc networks, thereby providing a foundation for understanding their applicability and the necessary design adaptations.

Embodied agents powered by large language models and vision-language models are increasingly deployed in physical environments, but jailbreak attacks can induce these agents to perform physically harmful actions. A growing number of guardrail methods have been proposed to intercept dangerous behavior before it is executed, yet existing safety benchmarks evaluate the embodied models themselves, leaving it unclear how well these guardrails actually defend an embodied agent in practice. We present the first systematic evaluation of jailbreak guardrails for embodied agents. To compare guardrails under identical conditions, we build a pluggable evaluation framework that treats the embodied agent as a fixed backend and each guardrail as a module that can intervene at the perception, planning, or control stage. We subject six representative guardrails to template-based and automated jailbreak attacks as well as safe instructions, and assess them at the system level along three dimensions: defense effectiveness, measured by the bypass rate and the hazard success rate in the simulator; usability, measured by the false-positive rate and the task completion rate on safe instructions; and efficiency, measured by the latency overhead added at runtime. Experiments on guardrails that span different intervention stages, decision mechanisms, and input modalities reveal a clear trade-off among the three dimensions, and show that no single guardrail dominates in all settings. We further analyze how intervention stage, decision mechanism, and input modality shape safety outcomes, and we offer practical guidance for selecting and designing guardrails for embodied agents.

Adverse weather and poor illumination remain major challenges for robust ego-trajectory planning in mobile autonomy. 4D radar offers reliable sensing under adverse conditions and direct radial-velocity measurements. However, sensing robustness does not necessarily translate into robust downstream planning, while existing 4D radar benchmarks focus primarily on perception rather than trajectory planning. We present Radar2Plan, a modular benchmark for open-loop ego-trajectory planning using real-world 4D radar data. Radar2Plan connects sensor encoders, scene representations, and planning heads through common interfaces, enabling controlled comparisons between different sensor and planner configurations. Using DSERT-RoLL and MAN TruckScenes, we evaluated seven combinations of camera, 4D radar, and LiDAR in four representative planning baselines and 12 weather and illumination conditions under a unified protocol. To our knowledge, Radar2Plan is the first benchmark dedicated to evaluating real-world 4D radars for autonomous driving planning. Experiments show that 4D radar alone supports competitive ego-trajectory planning and robust performance under adverse conditions. Sensor-configuration comparisons further demonstrate its complementary value to other modalities, while revealing dependencies on sensor combination and planning architecture. The modular design also supports additional datasets, sensing modalities, and planners, providing a flexible foundation for future research on radar-based autonomous driving planning.

Large learning rates can qualitatively change the trajectory of neural network training, often pushing optimization into regimes far from classical gradient-flow behavior. The Edge of Stability (EoS) offers a valuable lens on the dynamics such learning rates induce. We study corresponding dynamics in diagonal linear networks, where we uncover a competition between two distinct implicit biases that jointly determine the sparsity of the recovered solution in regression settings. Complementary to the Gain, which captures the average discretization error accumulated by Gradient Descent relative to Gradient Flow, we derive a closely associated but overlooked quantity: the Drift. Under large learning rates, it describes an imbalance between different discretization errors and represents a systematic shift in the optimization trajectory. While the Gain grows monotonically in certain regimes, and can bias towards denser, flatter interpolators, the impact of the Drift depends on its alignment with potential solutions, which can either counteract or reinforce the effect of the Gain. Consequently, its behavior drives model selection, particularly during early training epochs. To validate our theoretical insights, we introduce an intervention that actively steers the Gain to recover sharper, sparser solutions. Thus, our analysis reveals that large learning rates do not universally hinder the recovery of sparse solutions. On the contrary, they can be harnessed to control the implicit bias of training.

Quadratic-form test statistics are widely used in econometrics, and their performance depends on accurate variance estimation. Conventional plug-in estimators are consistent under the null hypothesis, but under alternatives the drift in the residuals inflates them and the test loses power. We develop a general framework for variance estimation in such statistics, replacing one of the two squared-residual factors by an auxiliary linear combination of the residuals (“cross-fitting”) chosen to annihilate the drift. We characterize the conditional bias of each estimator exactly. The drift enters the plug-in estimator squared, multiplied by quantities bounded away from zero, so its bias is positive whenever the drift is non-degenerate. It reaches the cross-fit estimator only through the part that survives the cross-fitting, and then only through off-diagonal entries of a residual-maker matrix. From this calculation we obtain conditions under which the cross-fit estimator remains consistent under alternatives while the plug-in estimator does not. At a common critical value the cross-fit test therefore rejects whenever the plug-in test does, and against distant alternatives the plug-in statistic converges to a finite limit, small when few observations carry the departure: its power can tend to zero where the cross-fit test's tends to one. We verify the conditions under primitive assumptions in nonparametric specification testing, overidentification testing, and testing many linear restrictions, and illustrate the procedure on the Oregon Health Insurance Experiment.

Preferential centrality (Hellervik, Nilsson and Andersson, 2019) models spatial activity through feedback between activity and attraction. We extend it with a density penalty. The local response depends on activity per unit of capacity and scales with capacity. A correction that is uniform per unit of capacity preserves total activity, and for a chosen response function and penalty strength, conservation determines this correction. We provide prefcent, an open-source reference implementation with convergence and feasibility checks and records of each run's inputs and settings. It reproduces the 200-iteration outputs of a companion Porto Alegre study across all 65,357 network segments, with median relative differences of order $10^{-14}$. In a strong-feedback example, the penalty reduces concentration and reaches the stopping tolerance in fewer iterations. Other synthetic examples show how penalty strength changes the relative prominence of centers and how their number varies with transport costs. A further example reaches different numerical equilibria from different initial states on the same network and capacity landscape. In multi-start sweeps on three synthetic two-dimensional landscapes, distinct converged states occur over a range of parameter values: at 172 of 420 points of a grid in feedback strength and transport cost, including 75 where the states differ in their number of centers. The note provides a model specification and a reproducible computational basis for further theoretical and empirical work.

Propagating systematic uncertainties from correction weights into templates generated from simulated data while preserving correlations in the signal extraction variables becomes increasingly challenging as analyses scale in size. SysVar's lightweight API hides most of the bookkeeping complexity of applying correction weights, building systematic variations, and histogramming nominal and varied templates. This streamlines workflows that are otherwise error-prone but essential for high-precision measurements. Correlations are preserved by propagating the full covariance across the analysis and compressing it into a small set of orthogonal eigenvariations, each defining an ordinary shape nuisance parameter. In this work, we demonstrate SysVar's workflow and impact by combining two independent pseudo-measurements that benefit from consistently correlated systematic uncertainties encoded in the template shapes. SysVar is open-source, pip-installable, and experiment-agnostic.

Modern LLM optimizers such as Muon often produce weight matrices with higher effective rank than Adam, yet further spectral control has delivered only modest gains. We identify a tension behind this result: concentrated spectra can suppress gradient directions in coupled weight matrices and slow optimization, while constraints maintained throughout training can limit task-specific adaptation and raise the attainable loss floor. We introduce ORCA (Orthogonal Regularization, Cooled After), a minimal optimizer intervention that applies strong but temporary soft orthogonality regularization early in training, then removes it. This allows the weights to benefit from a broader spectrum early on and adapt freely afterward. Across LLaMA, Qwen3, and fine-grained mixture-of-experts models ranging from 130M to 8B parameters, ORCA achieves lower final validation loss than Muon. Its loss reduction relative to Muon matches or exceeds Muon's reduction relative to Adam. Ablations support the early-shaping, later-release design. Further, ORCA requires no architectural changes and adds minimal overhead. An implementation is available at https://github.com/uuurraaa/ORCA_Optimizer.

Classical models such as smooth strongly convex minimization and strongly convex–strongly concave optimization combine two types of curvature information: curvature remains uniformly separated from zero, and its positive and negative orientations are prescribed in advance. Uniform nondegeneracy disentangles these roles, retaining the separation from zero while allowing the curvature orientation to vary across the domain. We identify a family of descent–ascent inequalities (DAI) that characterizes uniform nondegeneracy and naturally gives rise to Alternating Descent Ascent (ADA), a simple explicit method that alternates full-gradient descent and ascent. ADA converges globally and linearly to the unique stationary point from arbitrary initialization, with dimension-free first-order complexity. The framework recovers the smooth strongly convex setting as a limit and encompasses the strongly convex–strongly concave and linearly constrained settings, highlighting uniform nondegeneracy as a common first-order structure across various classical regimes.

Let $w \colon (0,1) \to \mathbb{R}_{+}$ be a weight. We prove that for an arbitrary Schauder basis $\{e^{i \lambda_n t}\}_{n \in \mathbb{Z}}$ in $L^2(0,1)$ and an arbitrary lacunary defect set $A \subset \mathbb{Z}$ the system $\{w(t) r_n(t)\}_{n \in \mathbb{Z} \setminus A}$ is always complete and never minimal for any weight $w$ satisfying a natural decay condition, which is sharp on the exponential scale. Moreover, we establish a simple combinatorial criterion for systems of the form $ \{ w(t) e^{2\pi i n t} \}_{n \in \mathbb{Z} \setminus A} $ to be complete and minimal in $L^2(0,1)$ for an arbitrary weight $w$ and a finite defect set $A \subset \mathbb{Z}$.

Quantum state preparation is a basic primitive of quantum computation and control, but its resource cost is usually analyzed for one target state or under an independent-copy construction. Here we study the many-copy problem under weighted Pauli control, where a Pauli-product rotation $e^{i\phi P}$ is charged by its rotation angle $|\phi|$, independently of the weight of $P$. We derive the exact single-qubit preparation cost and show that it is strictly nonadditive: a four-pulse collective protocol prepares two $T$ states at lower cost than two optimal one-copy protocols. We further identify finite regions of the Bloch sphere with collective saving, including trajectories that remain product throughout, showing that transient entanglement is not the only mechanism behind the advantage. Complementarily, a dimension-independent Rényi-entropy speed limit gives an extensive lower bound for every coherent target and an exponential fidelity penalty below a finite preparation rate. These results establish a many-copy geometry of state synthesis in which uncorrelated outputs can nevertheless benefit from collective control.

In high-dimensional factor pricing models, the power of regularized alpha tests depends on the ridge parameter, whose optimal value varies with the unknown direction of pricing errors. We address this tuning uncertainty by combining ridge-specific p-values using the Cauchy rule. The resulting test retains the least-squares alpha estimator and residual sample covariance and accommodates more assets than observations. We establish the joint Gaussian limits of the component statistics under the null and local alternatives and derive an explicit covariance formula across ridge parameters. These results characterize the combined test's asymptotic distribution and local power and justify its tail calibration. Simulations show empirical rejection rates close to the nominal levels and power gains over the fixed-ridge benchmark, with power approaching that of signal-informed ridge benchmarks across the simulated settings.

We consider a matching problem in which the cost of each edge is a vector with $k$ components. The cost of a matching is the sum of the bottlenecks over all components, and we ask whether there is a perfect matching of cost at most some value $Z$. This type of matching has applications in heavily synchronized job-shop scheduling problems and in reconfiguration problems, where movement is restricted to a single direction per step. In this paper, we analyze the problem from a parameterized complexity perspective and provide various results including FPT-membership for parameters $k$ and $Z$ combined, as well as W[P]-membership and W[SAT]-hardness for each of the two parameters individually. The reduction also implies para-NP-hardness parameterized by either maximum degree or treewidth. We further show hardness of approximation within a super-logarithmic factor for the optimization variant and provide a $k/d$-approximation algorithm for any constant $d \leq k$ as well as an efficient approximation scheme parameterized by $k$. With parameter $Z$, we show that no FPT-time $F(Z)$-approximation algorithm is possible for any computable function $F$, unless W[1] = FPT.

The realization of fractional Chern insulators has largely relied on isolated flat bands in complex bilayer or multilayer moiré systems with ideal quantum geometry. Whether such phases can emerge in genuinely dispersive bands, and by what mechanism, remains an open question. Here, we propose a mechanism for magnetic-field-induced strong correlations in the highly dispersive bands of monolayer graphene on hexagonal boron nitride. In the low-field classical-to-quantum crossover regime of the second moiré miniband, we show that the interplay of the moiré potential and weak magnetic fields (B < 1.5 T) profoundly alters electron dynamics. Saddle-point van Hove singularities produce extended open orbits that strongly suppress carrier propagation, potentially favoring a regime in which Coulomb interactions become comparatively more important, while weak magnetic fields lift valley degeneracy and trigonal warping redistributes Berry curvature. High-resolution magnetotransport and temperature-dependent measurements reveal fractional-slope transport features and signatures of interaction-driven incompressibility. Because the Hall conductivity is not fully quantized and the longitudinal resistance remains finite, we describe these observations as incipient FCI behavior: transport signatures consistent with a developing fractional Chern insulating state, without claiming a fully developed FCI phase established by complete Hall quantization and vanishing longitudinal resistance. Our results suggest a possible fermiology-driven route toward correlated topological states in dispersive bands in the classical-to-quantum crossover, distinct from the conventional flat-band scenario.

We establish a criterion for the irreducibility of Verma modules over symmetrizable Kac–Moody superalgebras. Our main tool is the Weyl-Kac denominator identity.

The lampshuffler group of $\mathbb{Z}$ is the semidirect product $\operatorname{FSym}(\mathbb{Z}) \rtimes \mathbb{Z}$, which consists of all permutations of $\mathbb{Z}$ that act as a translation outside a finite set. This infinite permutation group naturally contains as subgroups the wreath products $H \wr \mathbb{Z}$ for every finite group $H$. We prove that the Subgroup Membership Problem, and more generally, the Submonoid Membership Problem, are decidable in $\operatorname{FSym}(\mathbb{Z}) \rtimes \mathbb{Z}$. Our proof reduces Subgroup and Submonoid Membership in $\operatorname{FSym}(\mathbb{Z}) \rtimes \mathbb{Z}$ to Subgroup Membership in the wreath products $H \wr \mathbb{Z}$, which was shown to be decidable by Lohrey, Steinberg and Zetzsche (2015).

Compressive Synthetic Aperture Radar (SAR) imaging, typically formulated as an inverse problem and solved with traditional iterative optimisation methods, can be very computationally expensive. We investigate the use of denoising diffusion probabilistic models (DDPMs) for compressive SAR image reconstruction, where the diffusion model is guided by a poor initial reconstruction from sub-sampled data obtained via standard imaging methods. We augment this data-driven method with model-driven interleaved refinement processes, inspired by traditional compressed sensing (CS) methods, to enhance sparsity and heavy tails in SAR images. Experimental results on real ERS-1 SAR datasets demonstrate consistent improvements over baseline DDPM and state-of-the-art methods, with notable gains in reconstruction fidelity and structural similarity, while incurring only modest computational overhead. The proposed hybrid framework effectively combines the representational power and efficiency of diffusion models with the interpretability and robustness of model-based optimisation, enabling accurate compressive SAR imaging.

We study finite Laver algebras, the generalizations of Laver tables on sets of cardinality not of the form $2^n$. We characterize the conditions under which projections and embeddings between Laver algebras exist and show that there are $2^{\aleph_0}$ pairwise non-isomorphic direct limits of finite Laver algebras.

On-policy distillation (OPD) provides dense teacher supervision on student-generated trajectories, but generating and evaluating long rollouts incurs substantial training cost. Existing acceleration methods reduce this cost through open-loop rollout schedules or closed-loop horizon adaptation. However, supervision compatibility can vary substantially across trajectories, making a single rollout horizon difficult to match their heterogeneous reliable lengths: an overly short horizon may truncate useful supervision, while an overly long one wastes computation beyond reliable regions. Our key insight is that the trajectory-specific reliability boundary need not be predicted before generation. By viewing reliability as the first-passage of accumulated low teacher–student compatibility events, the boundary is inherently unknown before sampling, yet whether it has been reached can be determined exactly from the observed prefix. Building on this insight, we propose *Flash-OPD*, which shifts from rollout-horizon control to adaptive trajectory-level boundary verification. *Flash-OPD* interleaves cached generation with teacher verification and independently stops each trajectory according to its observed compatibility events. To reduce verification overhead, the recent event rate is used only to schedule the next verification point, while the actual stopping decision always relies on the exact cumulative count. This separation prevents estimation errors from causing premature termination while enabling efficient verification during generation. Extensive experiments across diverse datasets and teacher–student settings show that *Flash-OPD* achieves $2.2\times$–$7.5\times$ speedups over standard OPD while maintaining or improving accuracy.

Multimodal imitation learning requires diverse executable futures under the same observation and consistent behavior across replanning cycles. We present Conditional Trajectory Peaks (CTP), a single-pass policy framework that jointly predicts complete action-chunk candidates, probability masses, and trajectory scales. Distribution-Aware Peak Specialization (DAPS) specializes trajectory peaks using trajectory-level posterior responsibilities and mass- and scale-modulated overlap constraints. Evidence-Gated Trajectory Belief Transport (ETBT) maintains cross-chunk consistency through geometric correspondence between exchangeable candidate sets, while allowing current policy evidence to override historical constraints. CTP achieves a coverage score of 91.40% on Push-T; success rates of 100.0%, 79.72%, and 84.44% on D3IL Avoiding, Aligning, and Sorting-2, respectively. On LIBERO, CTP achieves an average success rate of 97.25%. In real-world dual-arm experiments, CTP preserves both placement modes in a two-plate task, succeeding in all 50 trials. On bottle uprighting and pen placement into a holder, it maintains success rates comparable to $\pi_{0.5}$ while reducing policy inference latency from 218.24 ms to 75.80 ms. These results demonstrate that single-pass trajectory modeling can combine multimodal behavior, closed-loop consistency, and efficient inference.

Risk-aware planners score paths on uncertain elevation maps using the path cost's mean and standard deviation. Modeling rigid contact, however, requires computing a maximum over several uncertain cells. First-order propagation loses accuracy here by differentiating at only a single cell, while Monte Carlo sampling requires a full path evaluation per draw. We compute the moments of that contact maximum in closed form using Clark's pairwise recursion. By tracking each contact's covariance against the shared map cells, we propagate the smooth remainder using exact Gaussian quadratic-form identities. A contest-depth calibration, fitted once on two design traverses, closes the aggregate standard-deviation shortfall that remains. On 317 held-out rover path segments, scored against a Monte Carlo reference from the same belief, every pre-registered criterion was met. The corrected Clark fold cuts the median error of the mean from linearization's 2.3% to 0.19% and attains the lowest error in the conditional value at risk (CVaR) at the 90% level of every method tested. A benchmark plan costs just 5.5 microseconds on a GPU. These accuracy gains concentrate at contested contacts. While they seldom change which path is chosen on this terrain, the fold still selects the reference-best path in 98% of decisions against linearization's 93 to 96%. On a second dataset the mean transfers, though the risk number does not.

We investigate CLIP for zero-shot gender estimation from full-face and periocular images. Three CLIP backbones are evaluated on 11,299 frontal images from Adience using image-text similarity with male/female prompts, achieving 95.54% full-face accuracy without task-specific training. For periocular, zero-shot predictions are strongly biased towards males, primarily due to a misaligned decision boundary. Threshold alignment substantially reduces this bias, reaching 85.29% accuracy. Linear SVMs trained on CLIP features provide only marginal gains, with a best periocular accuracy of 86.17%, approximately 2.8% above previous Adience results in the literature. Nevertheless, the gap with full-face performance confirms the greater difficulty of periocular gender estimation

Spatial prediction maps are crucial for environmental management yet highly sensitive to subjective modelling choices such as, e.g., covariates, spatial dependence structures, or mesh resolution. We develop a workflow that attributes the resulting map uncertainty to these choices and is applicable to any spatial model, since it operates only on an ensemble of maps produced by propagating the choices through spatial prediction. This ensemble is then filtered to retain only models with acceptable predictive performance, reduced in dimension by principal component (PC) analysis, and the sensitivity of the resulting scores to each choice is quantified by global sensitivity analysis based on optimal transport (OT) metrics. We demonstrate the framework with Bayesian geostatistics applied to the spatial mapping of sulphate concentrations in groundwater of Paris Basin (France), which is key for assessing groundwater chemical quality and establishing the natural geochemical baseline of aquifers. For this application, the covariates, mesh resolution, covariance models, and priors are combined into 18000 candidate configurations. These are filtered by cross-validated root mean square error (RMSE) and 95% coverage of the prediction interval, and the retained configurations are then inferred on a high-resolution grid and reduced to its three PCs. The OT analysis identifies covariate selection (under 95% coverage) and mesh resolution (under RMSE) as the dominant sources of uncertainty to the PCs, and a negligible one for the choice of the covariance model.

Realistic environment replicas are increasingly valuable for training and evaluating LLM agents, yet the original systems may be inaccessible or impractical to reproduce. We explore agentic language world modeling: rather than rebuilding an executable environment, a world model agent serves as the environment for a task agent and supports faithful and stateful simulation. We instantiate this paradigm with Trace2Env, a learning-free framework for settings where the original system is unavailable but historical interaction traces remain accessible. Trace2Env reconstructs these traces into a reusable environment worldbook containing environment schemas, grounded evidence, and induced behavioral knowledge. At runtime, the world model agent actively consults the worldbook together with persistent episodic state to infer each action's observation and lasting state effects. Across nine environments, Trace2Env improves both next-observation fidelity and long-horizon interaction consistency over conventional prompt-based LWMs. In multi-turn interaction, task agent actions generated against Trace2Env remain valid more often when replayed in the real environment, indicating that its simulated dynamics better preserve the consequences of earlier actions across successive turns. These results establish agentic language world modeling as an alternative direction for building realistic environment replicas without reconstructing the original executable system.