Papers

Conventional 3D Gaussian Splatting assumes a closed set of observations and long optimization schedules. Continual RGB-D mapping in contrast poses the problem that new observations arrive online, while previously reconstructed regions must be preserved. We present EliGSiR (Evidence-guided Load-adaptive Incremental Gaussian Splatting with Image Replay), a continual Gaussian mapper that controls how the available optimization budget is used as the reconstruction evolves. Map-Guided View Scheduling filters redundant incoming views and reconsiders retained views according to the current state of the map. Load-Adaptive Fidelity adjusts supervision resolution to the current mapping load instead of following a fixed resolution schedule. Targeted Geometry Growth separates depth supervision from Gaussian creation and adds geometric capacity only where repeated RGB-D observations indicate missing or misplaced structure. Together, these mechanisms adapt which views are optimized, how much image detail is used, and where the representation grows while mapping remains active. We evaluate EliGSiR on Replica, TUM RGB-D, ScanNet++, and real RGB-D sensor sequences, considering both the final reconstruction and the map available throughout acquisition. On TUM RGB-D fr3/long_office_household, EliGSiR reaches 21.52 dB with the same ground-truth mapping poses used by the controlled baselines, compared with 19.42 dB for SplaTAM. In the tracked-pose comparison, EliGSiR with live ORB-SLAM3 poses reaches 23.02 dB in 155.5 s, compared with 20.10 dB in 230.9 s for CaRtGS using its native tracker. We further evaluate reconstruction throughout acquisition and show how EliGSiR adaptive view scheduling, supervision fidelity, and geometry growth improve the use of the available mapping budget.

LEO satellite networks feature dynamic topologies, time-varying links, and diverse service requirements, which make conventional routing schemes difficult to support fine-grained quality-of-service (QoS) provisioning. Existing studies mainly optimize routing over network states with predefined objectives, but rarely address the practical challenge of translating unstructured natural-language service requests into adaptive routing decisions. To bridge this gap, we propose STR-Agent, an LLM-driven framework for QoS-aware routing in LEO satellite networks. The key innovation of STR-Agent lies in unifying intent perception, tool-based execution, experience accumulation, and reflection-based policy adaptation within a single agent architecture. Specifically, the Perception Module converts natural-language requests into structured routing semantics, while the Reflection Module dynamically adjusts the service-to-routing-policy mapping according to real-time congestion conditions and historical routing outcomes, rather than relying on a fixed routing objective. In addition, we develop a specialized perception model, and construct a domain-specific supervised fine-tuning dataset for LEO service understanding. Simulation results in a Walker-Delta constellation show that STR-Agent significantly outperforms conventional baselines: it reduces end-to-end delay by up to 60% compared with DQ-Dijkstra, improves average intent-understanding accuracy from 45.4% to 92.45% after supervised fine-tuning, and the Reflection Module further reduces the delay by 120 ms at 600 Mbps. These results demonstrate the potential of LLM-driven agent architectures to enable service-aware and adaptive QoS routing in future LEO satellite networks.

For the Dirichlet-type space $\mathcal D(\rho^{(1)},\rho^{(2)})$ on the unit bidisc $\mathbb D^2,$ where $\rho^{(1)}$ and $\rho^{(2)}$ are finite positive Borel measures on the closed unit disc $\overline{\mathbb D},$ we show that the Lévy measure $\nu_{(\mathscr M_z,1)}$ associated with the completely alternating multisequence $\left\{\|z^\alpha\|_{\mathcal D(\rho^{(1)},\rho^{(2)})}^2 \right\}_{\alpha\in\mathbb Z_+^2}$ admits the explicit representation \begin{equation*} d\nu_{(\mathscr M_z,1)}(x)=\frac{1}{1-x_1}d((S_{*}\rho^{(1)})\times\delta_1)(x)+\frac{1}{1-x_2}d(\delta_1\times (S_{*}\rho^{(2)}))(x) \end{equation*} on $[0,1]^2\backslash\{(1,1)\},$ where $S_{*}\rho^{(i)},$ $i=1,2$, denotes the pushforward measure of $\rho^{(i)}$ by the map $S:\overline{\mathbb D}\rightarrow [0,1]$ defined by $S(z)=|z|^2,$ $z\in \overline{\mathbb D},$ and $\delta_1$ denotes the Dirac measure at 1.

Enhanced performance of quantum reservoir computing has been associated with dynamical phase transitions, but whether this connection extends to dissipative systems and which relaxation mechanisms underlie it remain insufficiently understood. We systematically compare driven-dissipative Kerr reservoirs across first-order and continuous dissipative phase transitions and find enhanced memory and nonlinear processing near both phase boundaries. Although the closure of the Liouvillian gap marks the transitions in the thermodynamic limit, computational performance does not generally follow gap suppression. An exact post-training Liouvillian-mode decomposition quantitatively attributes the trained memory to intrinsic relaxation channels. It shows that the gap mode contributes only weakly, while finite-rate modes and their cross-contributions dominate the enhanced capacity. These results go beyond phenomenological correlations by directly linking memory capacity to the intrinsic Liouvillian relaxation spectrum. Moreover, these findings provide a physical basis for designing dissipative quantum reservoirs and testing memory-enhancement mechanisms in experimentally accessible Kerr platforms.

Chen and Xie recently proved, using the Ashikhmin–Barg lemma on minimal vectors, that every binary linear $[n,k,d]$ code with $k=n-2d+2+v$ ($v\ge 0$) has no codewords of weight in the interval $[2d-v,\,2d-1]$. Their argument uses two of the five basic properties of minimal vectors established by Ashikhmin and Barg (1998). In this note we utilize the third property, the disjoint-support decomposition of non-minimal codewords in binary codes, to generate a mirror vanishing band on the other side of $2d$: if $A_{d+1}=\cdots=A_{d+t}=0$ for some $t\ge 1$ and $k\ge n-2d+1$, then $A_w=0$ for all $w\in[2d+1,\,2d+t]$. Combining the two bands, the number of nonzero weights of such a code is at most $n-d-v-2t+1$, improving the Chen–Xie bound $n-d+1-v$ by $2t$.

Conventional bandwidth-limited ranging resolves delay on the scale of \(c_0/(2B)\), yet a large distributed aperture can support substantially finer localization through near-field carrier-phase diversity. This letter characterizes the coherent main-lobe width of the localization likelihood for pinching-antenna systems and shows that it is governed by the spread of the direction cosines subtended by the aperture at the target. The resulting expression recovers classical far-field angular and Fresnel-range scalings and reduces to a wavelength-scale cell when the aperture subtends a large solid angle. Guided by this characterization, a hierarchical maximum-likelihood search is developed, combining noncoherent coarse localization, coherent lobe selection, and local refinement. Numerical results under scattering demonstrate submillimeter localization above a threshold signal-to-noise ratio without localization failures in the tested ensemble.

This paper establishes the existence of three distinct weak solutions for a class of nonhomogeneous quasilinear inclusion systems, which are governed by locally Lipschitz functionals within Orlicz-Sobolev spaces over unbounded domains $\mathbb{R}^{N}$.By employing a new three critical points theorem established by Wu-Zhou in [29], we extend the nonsmooth critical point theory to handle scenarios with nonlinear growth conditions.Our analysis imposes a precise structural condition on the nonlinear term $H$, requiring it to exhibit superlinear growth while maintaining subcritical asymptotic behavior. This condition establishes a delicate balance between growth restrictions and functional analytic requirements. The novelties of this work include the construction of suitable energy functionals that are consistent with both the structure of Orlicz-Sobolev spaces and the characteristics of nonlinear growth. Through rigorous variational analysis and careful estimation techniques, we establish the existence of three weak solutions under these generalized conditions, thereby significantly expanding the applicability of critical point methods in nonstandard function spaces.

Magnetic Resonance Imaging (MRI) acquired at different field strengths exhibits pronounced variation in noise, resolution, homogeneity, and contrast, which limits comparability across acquisition settings and complicates downstream analysis. We address this with a unified conditional model for controllable field-to-field synthesis, built on the framework of conditional latent bridge matching. Our single model achieves highly competitive results across the validation phase for all three tasks of the MRIxFields2026 challenge without task-specific architectures or training. We achieve fast generation with only a single inference step, producing all modality and field-strength combinations for $30$ axial slices in under $90$ seconds, as well as cross-modality-strength translation for a full volume in under $70$ seconds, on a single NVIDIA A5000 GPU. We further provide extensive ablations regarding different components of our solution. Code: https://gitlab.com/siddharthsrivastava/mrixfields-2026

Ultrasound image analysis plays a crucial role in cancer screening and prenatal diagnosis, yet comprehensive assessment requires jointly addressing tasks such as lesion segmentation and benign-malignant classification. While recent vision foundation models have shown remarkable universal representations, unlocking their potential for ultrasound is bottlenecked by the considerable domain gap from natural images. Existing methods typically fine-tune heavy vision encoders for isolated tasks, incurring substantial computational overhead while overlooking the underlying commonalities across heterogeneous tasks. In this work, we propose FreqDINO++, a frequency-guided multi-task routing vision foundation model for universal ultrasound analysis. We first introduce a Multi-task Routing Adapter (MR-Adapter) to support parameter-efficient integration of task-common and task-specific knowledge, a Frequency-aware Feature Enhancer (F$^2$-Enhancer) is then designed to capture the rich multi-scale frequency characteristics of ultrasound images, and a Task-aligned Collaborative Decoder (TC-Decoder) is devised to promote collaboration between dense and global prediction tasks through global-local token interaction. Extensive experiments on large-scale multi-task and external single-task ultrasound benchmarks demonstrate that FreqDINO++ consistently outperforms strong baselines and recent foundation models across 27 diverse clinical task scenarios, while also showing promising generalization to unseen data. The code is at https://github.com/MingLang-FD/FreqDINO-Plus.

We analyse the essential spectrum of ${\mathcal M} = {\operatorname{curl}} {\operatorname{curl}}$ acting on divergence-free vector fields in unbounded domains of $\mathbb{R}^3$. We show that $\sigma_e({\mathcal M}) = [0,+\infty)$ in quasi-conical domains and $\sigma_e({\mathcal M}) \neq \emptyset$ in quasi-cylindrical domains. For horn-shaped domains with circular cross-section and eventually mean-convex boundary, we establish that $\sigma_e({\mathcal M}) = \emptyset$, independently of their volume. For horns with annular cross-section, the transverse normal harmonic field determines an effective one-dimensional Schrödinger operator $H_V$ with $\sigma_e({\mathcal M}) \supseteq \sigma_e(H_V)$. Finally, for a concrete family of perforated exponential horns with a double-exponential hole, we show that depending on the rate of shrinking of the hole at infinity, either $\sigma_e({\mathcal M}) = \emptyset$, or $\sigma_e({\mathcal M}) = [\gamma^2, +\infty)$, or $\sigma_e({\mathcal M}) = [0,+\infty)$.

We study how a close passage past an intermediate-mass black hole (IMBH) changes the evolution of an eccentric white dwarf (WD) binary that has nominal Roche lobe overflow at internal pericenter. Using three-dimensional smoothed particle hydrodynamics, we follow binaries with eccentricity $e_{\rm in}=0.6$ and mass ratios $q=0.2$ – $0.7$ on parabolic, retrograde orbits around a Schwarzschild IMBH at two encounter strengths, $\beta^b=0.5$ and $1.5$. We compare them with the same binaries evolved in isolation. The IMBH greatly speeds up the merger. Every IMBH-perturbed binary that merges does so during or shortly after its second internal pericenter passage, which is always less than the time taken to merge in isolation. The merger outcome is not monotonic with encounter strength. The $q=0.2$ binary retains two bound cores for $\beta^b=1.5$ but merges at $\beta^b=0.5$. In the deeper encounter it settles onto a less eccentric post-encounter orbit, showing that the timing of the encounter relative to the inner orbit plays an important role. Deep encounters also produce transient reversals of the internal orbital angular momentum for $q=0.2$ – $0.4$, while the material bound to the accretor gains more spin angular momentum than that bound to the donor. The specific orbital angular momentum of the binary CM around the IMBH changes by only a few $\times10^{-4}$ of its initial value. We also identify He-origin material that can burn helium on a dynamical timescale, amounting to $\sim15$ – $17\%$ of the He-origin mass for $q=0.7$. IMBH encounters can therefore quantitatively change the evolution of eccentric WD binaries.

In this note, we give a simple analysis of a non-adaptive simulated annealing algorithm for estimating the partition function of Gibbs distributions. This yields the most efficient reduction of this kind so far. We also establish lower bounds for both general and non-adaptive algorithms, showing that our algorithm is optimal over a broad range of parameters.

Deceptive online job advertisements have emerged as a primary pathway into forced labour, yet systematic detection methods remain underdeveloped due to data scarcity and absence of empirically validated indicators. We formalise this detection challenge as a classification problem under signalling theory, where exploiters transmit costless signals mimicking legitimate communications across textual, visual, and structural dimensions. Using 464 verified cases (164 deceptive, 300 legitimate) collected through anti-slavery charities across nine origin countries and 21 industries, we develop multimodal detection models combining computer vision, natural language processing, and semantic embeddings. Through systematic feature ablation experiments and repeated stratified cross-validation, we demonstrate that individual modalities achieve substantial discriminatory power (ROC-AUC: 0.87–0.97), whilst their integration yields modest further gains. SHAP-based analysis reveals that text quality and domain-specific risk language are the primary discriminators, with readability indices, risk keyword density, and visa sponsorship mentions ranking highest, followed by visual colour and texture features. These production quality gaps reflect resource constraints that prevent exploiters from maintaining professional standards across all communication channels simultaneously. We operationalise findings through a proof-of-concept decision support system providing interpretable risk scores for practitioners. This work demonstrates how rigorous analytical frameworks can address complex humanitarian operations challenges characterised by information asymmetry and limited ground-truth data.

Transformers have emerged as the dominant architecture in sequence modeling, achieving remarkable success in natural language processing and reasoning tasks. While existing literature has established the Turing completeness of transformers under bounded input length, the reasoning power of a single transformer operating on inputs of unbounded length is not fully explored. In this paper, we theoretically investigate the reasoning limitations of a single transformer and the enhanced capabilities of agent systems. We show that a single fixed finite precision transformer cannot memorize certain Turing machines with inputs of arbitrary length, such as the arithmetic; and a single fixed infinite precision transformer trained with a random algorithm is not Turing complete with probability one under reasonable conditions. To overcome the limitation of a single transformer, we define a formal agent architecture consisting of decision, execution, and memory modules and show that for any Turing machine $\mathbb{T}$, there exists an agent that can memorize $\mathbb{T}$ and is computationally the same as $\mathbb{T}$. Thus, agents are Turing complete.

Traffic regulations are written for human interpretation and therefore rely on shared background knowledge and flexible phrasing, which inherently introduce ambiguity, context dependence, and semantic underspecification. These linguistic characteristics conflict with the precision required by computational reasoning engines such as Prolog, which demand explicit logical structure. This study evaluates two baseline translation approaches, Natural Language to Prolog ($NL\rightarrow Prolog$) and Logical English to Prolog ($LE\rightarrow Prolog$), and introduces a new reasoning-guided translation framework called Structured Four-Stage Legal Translation ($S4L\rightarrow Prolog$). The proposed S4L framework performs semantic role extraction, scene completion, logical mapping, and Prolog rule generation within a single guided prompt, enabling direct translation of raw traffic rules into executable logic without human intervention. A benchmark consisting of twenty real-world traffic rules was used to evaluate each approach in terms of syntactic validity, semantic correctness, and logical completeness. $S4L\rightarrow Prolog$ achieves the highest accuracy, correctly formalizing 75 percent of the rules, while $NL\rightarrow Prolog$ reaches 60 percent and $LE\rightarrow Prolog$ reaches 55 percent. Qualitative analysis further shows that S4L captures implicit causal relations, deontic modality, and exception structure more reliably than the baselines. These results demonstrate that structured reasoning prompts can substantially improve the reliability of natural-language-to-logic translation for legal and safety-critical applications.

We study reconstruction in autoencoders that apply the same forward map before and after setting the observed coordinates to zero. For equal odd input and hidden dimensions $d\geq 3$, among orientation-preserving diffeomorphisms whose Jacobian singular values lie in $[m,M]$, we show that the least uniform reconstruction-derivative error is $\max\{1-M(M-m)/2,0\}$, with affine maps attaining this sharp bound at every prescribed depth. A translated radial rotation can nevertheless reconstruct any prescribed ball exactly with singular values arbitrarily close to one, motivating additional conditions for a finite-data bound. We test this prediction on a 798,452-point terrestrial LiDAR forest scan. At input scale $0.05$, the mean theoretical bound is $0.155$, about $84\%$ of the mean normalized training error $0.185$ across four spatial regions, two depths, and three seeds. At this scale, adding one hidden coordinate reduces the mean reconstruction error below $6\times10^{-6}$.

We determine the branching of irreducible unitary representations of the basic classical Lie superalgebras $\mathfrak{sl}(m|n)$, $\mathfrak{osp}(m|2n)$, and $F(4)$ under restriction to their even subalgebras. The proof combines Dirac inequalities for the Huang-Pandžić-Dirac operator $\operatorname{D}_{\mathfrak{g},\mathfrak{g}_{\bar{0}}}$ with a super analogue of Freudenthal's recursion derived from the Dirac formalism. We extend the Dirac method to positive systems that are not adapted to the relevant real form by transporting the Dirac inequality through odd reflections and determining the resulting correction terms. This yields finite recursive procedures that determine the occurring even constituents and their branching multiplicities, together with closed formulas for the latter, uniformly in the degree of atypicality. For $\mathfrak{sl}(m|n)$, the corrected Dirac inequalities also give a characterization of unitarity for all positive systems relevant to unitary representations.

Many signal processing tasks require intensive computations. Quantum computing promises to accelerate certain tasks, but algorithms must be designed around the limitations of quantum mechanics. This paper provides a quantum implementation of finite impulse response (FIR) filters, which are a widely used tool in classical signal processing. The filter can be parallelized and composed as part of larger quantum algorithms, with potential for speedups using quantum amplitude estimation and future fault-tolerant quantum hardware. To accomplish these properties, we introduce a state-space framework wherein sample-based signal processing is performed with unitary quantum circuits. We present the quantum delay gate as the quantum analog of the delay line in discrete-time signal processing. Unitary circuits intrinsically describe lossless systems, but we also implement lowpass and other bounded filters by emulating a projection operator. The FIR filter implementation is tested and demonstrated with a quantum circuit simulator.

Blind and low-vision users often need to locate a specific personal object rather than an arbitrary instance of the same category. The task calls for a robot that can move through the space and reach viewpoints the user cannot, and for an accessible interface where the user says which object is meant and learns whether the right one was found. We present RoboFind, a multi-agent framework in which a smartphone teaches the target and a quadruped robot carries out the search. A Target Teaching Agent converts guided smartphone recordings into a semantic target profile and a reusable multi-view reference bank through an accessible capture flow with AR guidance, speech and haptic feedback, and screen-reader support, so later missions refer to a stored object without repeating the teaching process. At runtime, a Navigation Agent explores the environment and proposes candidate targets, a Verification Agent checks each candidate against the stored references, and a Coordination and Recovery Agent completes the mission or triggers recovery and continued search. Across 32 real-robot missions, RoboFind reaches 85.0% success against 25.0% for a reconstructed sequential first-stop baseline over 20 trials with ten targets, and reduces false success from 75.0% to 5.0%. On six shared targets it succeeds in 10/12 trials, against 5/12 for 12 independently executed GPT-6 Astra-only trials. These results show that the multi-agent design fits the demands of personalized object search, where verifying object identity before declaring completion is what makes the outcome something a user can rely on.

Primordial black holes (PBHs) formed in the early universe represent a well-motivated dark matter candidate. However, their abundance in the intermediate-to-high mass range ($10^6$-$10^9\,M_\odot$) remains relatively less constrained. In this mass regime, the corresponding Einstein radii on the milliarcsecond scale make Very Long Baseline Interferometry (VLBI) the most direct observational probe via milli-lensing of compact radio sources. In this paper, we present predictions for the abundance of PBH dark matter ($f_{\rm PBH}$) using the large SMILE sample of $\sim 5000$ flat-spectrum compact radio sources. Under the optimistic observational conditions (with an angular resolution of $1.5\,\mathrm{mas}$, a maximum image separation of $100\,\mathrm{mas}$, and a maximum detectable flux ratio of $40$), assuming that no milli-lensing events are confirmed, we obtain the projected 95% upper limit of $f_{\rm PBH}\lesssim0.12%$ in the intermediate-to-high mass range, which would be approximately an order of magnitude tighter than previous VLBI milli-lensing limits. Broader lognormal mass functions yield correspondingly weaker projected limits. Generalising the optical depth to account for magnification bias and finite-core resolvability changes these projected limits by only about $5$%. Moreover, we show that the predicted limits are highly sensitive to a small number of confirmed events, i.e., even one confirmed milli-lensing event would relax the projected upper limit to $0.20$%, highlighting the critical importance of robust candidate confirmation in the SMILE sample. Finally, under the optimistic baryonic accretion scenario, the PBH mass growth changes the projected limits by $\lesssim 1$% within the VLBI-sensitive mass range, confirming the robustness of our results.

The application of microfabricated nanocalorimeters for real-time, highly sensitive, and selective monitoring of the energy fluxes from low-pressure plasma discharges has been reported recently. Here, we describe the experimental setup for complementary characterization of nanocalorimeters performance using plasma diagnostics techniques. Langmuir probe, retarding field energy analyzer (RFEA), optical emission spectroscopy (OES), quadrupole mass spectrometry (QMS), and thermocouple-based thermometry constitute the plasma monitor suite. We discuss two application examples of the plasma monitor suite: (1) discrimination between electron- and ion-induced heat fluxes via sensor direct current (DC) biasing, and (2) detection of fast temperature transients in pulsed discharges. The main goal of this work is to demonstrate a nanocalorimetry platform as a real-time metrology for plasma processing and process control.

We study smooth strongly convex–strongly concave minimax optimization in the deterministic unconstrained setting, without assuming a bilinear or separable structure. Although existing multi-loop methods attain near-optimal condition-number dependence, standard single-loop methods generally exhibit a substantial complexity gap. To close this gap, we propose the Single-Loop Predictor–Corrector Extragradient Method with Damped Momentum (PCE-DM), which combines an extragradient prediction–correction scheme with a novel auxiliary feedback recursion for the weaker-curvature variable. PCE-DM uses fixed parameters and two new full-gradient evaluations per iteration after one initialization query, while requiring no inner solves, accuracy schedules, or staged restarts. We develop a Lyapunov analysis that controls the predictor–corrector mismatch through corrected-gradient increments and establish last-iterate linear convergence. Specifically, PCE-DM computes an $\varepsilon$-accurate relative solution, measured by the squared Euclidean distance to the saddle point, within $\mathcal{O}\!\left(\sqrt{\kappa_x\kappa_y} \log(2\kappa_x\kappa_y/\varepsilon)\right)$ full-gradient queries. This result closes the condition-number complexity gap between standard single-loop methods and near-optimal multi-loop methods, matching the known lower-bound order up to logarithmic factors while retaining fixed, explicit single-loop updates. Numerical experiments on regularized linear regression and AUC maximization demonstrate the computational efficiency of PCE-DM.

Sequential thermal evaporation (sTE) is emerging as a solvent-free route to high-quality mid-bandgap perovskites, but its extension to mixed-halide wide-bandgap (WBG) and Sn-Pb narrow-bandgap (NBG) absorbers for all-perovskite tandem solar cells (TSCs) remains limited by an incomplete understanding of solid-state conversion. Here, using time-sliced ex situ analysis, we reveal divergent solid-state conversion mechanisms in sequentially evaporated WBG and NBG precursor stacks. In WBG stacks, formamidinium (FA)-containing species penetrate the inorganic template and Br/I redistribution precedes substantial three-dimensional perovskite formation. The photoactive phase then crystallizes from a chemically mixed reservoir, and absolute PbBr$_2$ thickness, rather than nominal PbBr$_2$/PbI$_2$ ratio, determines the final bandgap. In NBG stacks, by contrast, an early Pb-rich perovskite phase forms upon formamidinium iodide deposition, restricting further FA penetration into the buried SnI$_2$ precursor. Subsequent annealing promotes rapid lattice reorganization faster than Sn/Pb interdiffusion, leaving vertical compositional gradients. Guided by these insights, we develop sTE absorbers with bandgaps spanning 1.26-1.96 eV and demonstrate the first evaporated all-perovskite TSC, reaching a power conversion efficiency of 19.2%. Encapsulated tandems retain on average 80% of their initial efficiency after 1,200 h at 65 $^\circ$C (ISOS-D-2). These results establish bandgap-specific control of solid-state conversion as a design principle for sequentially evaporated perovskite tandem photovoltaics.

Agricultural image understanding requires fine-grained recognition of plant diseases, pests, crop structures, and botanical species under complex real-world conditions. Despite recent advances in Multimodal Large Language Models (MLLMs), existing models remain limited to text-only outputs and lack pixel-level visual grounding capabilities. In this work, we introduce AgriScope, a unified pixel-grounded multimodal framework for agricultural image understanding. AgriScope jointly supports image-level, region-level, and pixel-level understanding within a unified framework, enabling tasks such as grounded caption generation, referring expression segmentation, and multi-turn multimodal interaction for agricultural imagery. AgriScope integrates biologically specialized semantic representations with dense spatial grounding through biological-semantic encoding, dense spatial representations, and pixel decoding. To support large-scale grounded learning, we introduce AgriGround, a large-scale pixel-grounded agricultural multimodal instruction-tuning dataset containing over 500K images and 11M instruction-following samples spanning plant disease analysis, crop and weed identification, insect pest recognition, and fine-grained botanical understanding. AgriGround is constructed through a multi-stage automatic annotation pipeline that integrates multimodal caption generation, phrase-level grounding, segmentation mask generation, and task-oriented instruction synthesis to produce densely grounded supervision. Extensive experiments across multiple agricultural vision-language tasks demonstrate the effectiveness of AgriScope in pixel-grounded multimodal understanding, establishing a strong benchmark for agricultural vision-language learning and visual grounding. The dataset and code will be made publicly available at (https://github.com/boudiafA/AgriScope)

A cell's identity depends on where it sits in tissue: for example, a macrophage behaves differently in a tumor core than at its edge. Spatial transcriptomics has transformed how we study this by recovering that lost coordinate, but it does so by producing data that is simultaneously high-dimensional, multimodal, and uncertain. Visualizing this combination is a hard problem in its own right, and one that warrants an assessment of how the field currently represents it, what has worked, and what is still missing. We surveyed 148 papers and 1,824 figure panels using a What-Why-How coding framework grounded in Munzner's nested model, connecting the data represented, the biological tasks motivating each visualization, and the design choices through which they are expressed; a subset of the surveyed work also contributed dedicated interactive visualization software that was not necessarily reflected in the static figures, and we looked at what interaction capabilities those tools supported as well. We close by outlining where the field stands and the challenges ahead for bioinformatics and visualization researchers to tackle together.

Three-dimensional reconstruction from unorganized point clouds remains a challenging problem in computer vision, geometric modeling, and computer-aided design. While neural implicit methods achieve impressive reconstruction accuracy, geometry is typically encoded in latent representations that limit interpretability and reuse within engineering workflows. We present NeuSOGA3D (Neuro-Symbolic Observation-Guided Geometric Abstraction in 3D), a hybrid framework that combines learned perceptual priors inherited from NeuSOGA with explicit symbolic geometric reasoning. The method projects point clouds onto principal orthographic planes, constructs symbolic implicit spline representations from the resulting observations, and fuses them through shape-preserving constructive solid geometry operations to generate a coarse visual hull. Additional geometric detail is recovered through cross-sectional decomposition and volumetric reconstruction using Partial Shape-Preserving Splines. Unlike conventional neural implicit approaches, NeuSOGA3D progressively transforms observations into explicit symbolic entities, including control polygons, implicit spline fields, cross-sections, and volumetric lofts. Experiments on all forty categories of the ModelNet40 benchmark demonstrate the ability of the framework to recover structurally meaningful and CAD-compatible geometric representations from diverse point-cloud observations. The results highlight the potential of combining learned perception with symbolic geometric reasoning for explainable geometric intelligence.

Recent crises have made supply security central to trade policy. We show how tradeable import certificates (TIC) implement targets for domestic production and reliable foreign supply while preserving gains from trade. A single certificate market per country decentralizes the welfare-maximizing allocation under heterogeneous targets, with certificate prices adjusting endogenously as conditions change. TIC robustly protect these targets against a range of deviations from trade agreements. A model of economic coercion microfounds the targets, linking them to shortage salience, supply reliability, bilateral trade dependence, and norms against yielding to coercion. Tariff-subsidy agreements require more information and are more vulnerable to hidden deviations. While a common carbon price provides a natural focal point for climate agreements, quantity-based security targets could be a more natural focal point for trade agreements than tariff and subsidy rates.

Time series of random objects, such as covariance matrices, probability distributions, and functional data, call for forecasting methods that do not rely on standard arithmetic operations. We introduce geodesic exponential smoothing, a generalization of exponential smoothing to time series in Hadamard spaces: the forecast level moves a fixed fraction of the way along the geodesic toward each new observation. The smoothing parameter is estimated by minimizing the average squared distance between observations and their forecasts. We further introduce an innovations mechanism under which each observation has conditional Fréchet mean equal to the current level, providing the metric-space analog of the innovations state-space model. In contrast to autoregressive models for object-valued time series, the framework involves a single scalar parameter, assumes no stationarity, and updates online in constant time per observation. Under this mechanism, we establish sample-path properties of the generative process via the quasilinearization available in Hadamard spaces, and prove almost-sure consistency of the smoothing-parameter estimator. Three real-data applications, spanning covariance-matrix, distributional, and functional time series, assess the forecasting performance of the method against structurally heavier alternatives.

Image quality assessment (IQA) datasets use different subjective protocols and rating scales, so their scores are not directly comparable. The lack of a common perceptual scale hinders multi-dataset training and precludes direct inter-dataset evaluation. We address this by conducting a new subjective experiment and using its ratings as perceptual anchors to fit monotonic mappings that place the existing scores of 23 IQA datasets on a common quality scale while preserving within-dataset rankings. The resulting dataset, MOSAIQ-500K, contains over 500,000 images and is, to our knowledge, the largest IQA dataset with perceptually aligned subjective scores. We also propose MOSAIQ-Bench, an inter-dataset benchmark, and use it to evaluate 31 IQA methods, revealing substantial gaps between intra- and inter-dataset performance, particularly on authentic distortions. The aligned scores offer a key advantage: replacing specialized multi-dataset training mechanisms with standard regression losses yields comparable intra-dataset accuracy and better inter-dataset performance. MOSAIQ thus provides a practical foundation for combining heterogeneous IQA datasets to train more generalizable IQA models. Code and dataset will be made public at ivc.uwaterloo.ca/projects/unifying_iqa_datasets/ upon acceptance.

Speculative Decoding (SD) has significantly accelerated Large Language Model (LLM) inference, yet existing approaches face a fundamental tradeoff between two drafting strategies: neural drafting and context-based copying. Neural drafts (e.g., EAGLE3) provide robust performance across diverse text settings, while copy-based methods achieve higher speedups in copy-intensive regimes by generating candidates faster and exploiting long repetition spans for near-perfect speculation. We analyze existing copy-based methods and find that they are prone to accidental repetitions where surface-level n-gram overlap does not reflect a structural intent to copy, leading to false-positive triggers that ultimately degrade throughput. We introduce SwitchSD, an adaptive framework that treats copying as a latent control signal of the LLM. By training lightweight probes on the target model's internal representations, SwitchSD identifies genuine copy-intent with high precision (AUC > 0.99). This allows the system to dynamically switch between neural drafting (e.g., EAGLE) and context-based copying. Our results across Llama and Qwen families demonstrate throughput gains of up to 15% over state-of-the-art baselines like EAGLE3, effectively turning copying from a noisy heuristic into a principled, model-aware decoding regime.