Papers

For a graph \(G\) admitting a real symmetric realization with exactly two distinct eigenvalues, \(MB(G)\) is the minimum, over all such realizations, of the smaller of the two eigenvalue multiplicities. Adm, Fallat, Meagher, Nasserasr, Plosker, and Yang asked for this parameter for the complement of a path on at least eight vertices. We answer their question completely by proving $$ MB(\overline{P_n})=3 \qquad (n\ge 6). $$ In particular, this resolves the previously unresolved orders \(n\ge 9\) divisible by three. The proof is exact and constructive. We exhibit six vectors in \(\mathbb{R}^3\) whose mutual inner products vanish exactly for consecutive indices, whose rank-one outer products form a basis of \(\mathbb{S}^3\), and which admit a strictly positive Parseval scaling. An elementary absorption lemma then permits any finite faithful orthogonal extension of this vector chain to be added with small positive weights while the six original weights are corrected to retain the Parseval identity. The resulting Gram matrix is a rank-three orthogonal projection in \(\mathcal{S}(\overline{P_n})\). A local two-dimensional orthogonality obstruction gives the matching lower bound. For completeness, we include self-contained proofs of the exceptional small orders: \(MB(\overline{P_3})=1\), whereas \(q(\overline{P_4})=4\) and \(q(\overline{P_5})=3\).

We introduce a self-supervised framework for learning joint visuospatial representations from RGB-D observations. While modern vision foundation models are trained almost exclusively on RGB images, many embodied systems have access to explicit depth sensing, which provides geometric information that monocular inputs cannot recover. Our method integrates depth-derived geometric priors with a visual backbone through inter-patch and intra-patch fusion, enabling the model to encode both appearance and spatial structure efficiently. The resulting representation shows promising improvements on 3D awareness while preserving semantic transfer: it outperforms prior methods of comparable scale on multiple 3D geometry benchmarks, and remains competitive when probed for standard RGB-D semantic segmentation tasks.

Effective human-robot collaboration in industrial settings requires robots to understand human intentions and assist with task planning, reducing workload. Recent works have explored the use of Multi-modal Large Language Models (MM-LLMs) for task planning in such data-scarce scenarios, leveraging in-context learning to interpret user actions and generate long-horizon action plans in natural language. However, MM-LLMs inherently lack an understanding of system states and do not track state transitions, often leading to hallucinated actions that deviate from the intended goal. Additionally, generating action plans in natural language tends to limit the generated plans to a high level, introducing ambiguity in action execution. To address these limitations, we propose the State-aware Task Estimator and Planner (STEP), which prompts a MM-LLM to explicitly estimate the state of the system and predict the state transitions resulting from executed actions. By forecasting future states alongside actions, STEP ensures task-convergent planning while also providing additional assistance parameters necessary for executing the predicted actions. We evaluate STEP in a simulated environment using a robot assembly task. Our approach outperforms the state-of-the-art by 32.8% in action executability and 14.8% in final-state error.

Expansion planning models usually fix storage duration before optimization, setting how much storage power to build but not for how long it can discharge. We present a generation-transmission-storage expansion framework in which candidates of many durations compete on annualized cost, making the duration mix an optimization outcome. It is a rolling-horizon, two-stage stochastic linear program, each five-year stage is a full year at hourly resolution under ten coherent inflow, wind, and solar scenarios. We apply it to the Brazilian Interconnected System over 2030-2050, where demand roughly doubles to 1,716 TWh/year and variable renewable energy (VRE) supplies most new capacity. Two cases are compared: 36 pumped-storage hydropower (PSH) candidates over four subsystems and nine durations (4-144 h) alongside 4-h battery energy storage systems (BESS), and BESS alone. With PSH available, the model builds 31.2 GW / 755 GWh of PSH and no BESS, dominated by 12-h capacity with 4.9 GW of 72-h units. Without PSH it builds 38 GW / 152 GWh of BESS and 7.8 GW more gas-fired capacity, mostly open-cycle peakers. PSH lowers 2050 annualized system cost by US$ 5.0 billion/year (6.8%), operating cost by 23.5%, thermal generation by 34 TWh/year, and long-run marginal cost by 20%; 2035 VRE curtailment falls from 8.4% to 3.2%. The magnitudes are specific to Brazil, but the underlying mechanism is general: limiting storage candidates to a single duration understates the system's optimal energy-storage requirement and overstates its residual need for thermal capacity.

Robots that can change their morphologies and behaviors for different tasks and environments hold great promise for adaptable, multifunctional systems. Modular reconfigurable robots (MRRs) can achieve such functionalities by docking and rearranging individual units, but most rely on rigid modules that lack structural compliance, resulting in limited capabilities. Continuum robots offer compliance through flexible backbones, yet they cannot self-reconfigure into task-adaptive multi-robot configurations. Here, we introduce an MRR that unifies the advantages of both architectures by combining a tensegrity-based compliant body with claw-based connection mechanisms. Each robot can manipulate and locomote independently, and multiple robots can self-reconfigure into different morphologies (e.g., chains, loops, branches) for cooperative manipulation and locomotion. We demonstrate the robots' capability across diverse tasks and environments, including coordinated object manipulation and transport, multimodal locomotion, and loco-manipulation in real-world scenarios. These results lay a foundation for adaptable and multifunctional robotic collectives, with broad potential applications in manufacturing, space exploration, and search-and-rescue operations.

Mental health assessment relies on episodic self-report scales, which convert subjective states such as stress into numerical scores but provide only sparse snapshots of wellbeing. Wearable devices offer longitudinal behavioral and physiological signals for continuous, low-burden monitoring. Recent LLM-driven personal-health agents enable natural language queries over wearable signals, but mainly handle short-term, retrieval-based lookups (e.g., highest step count over a week). They do not evaluate whether agents can reason over long-term signals to predict wellbeing scores paired with evidence-grounded rationales. To address this gap, we introduce BALMS, the first systematic benchmark of LLM-based agentic systems for longitudinal mental health sensing. BALMS spans 3 real-world longitudinal datasets, 2 task families (closed-form wellbeing-score prediction and rationale generation auto-graded by an LLM-as-Judge), 3 agentic paradigms evaluated across 5 open- and closed-source LLM backbones. We find that zero-shot agents rarely outperform a simple mean baseline, except with stronger backbones or compact, semantically meaningful features. Chain-of-thought prompting improves reasoning-oriented backbones, but does not guarantee temporal grounding or numerical correctness. Together with more analysis on efficiency and temporal scaling, BALMS highlights the need for longitudinal mental health agents that selectively retrieve history, ground temporal evidence, and reason over interpretable behavioral features.

In this paper, we develop a deterministic, physically constrained generative framework based on the Madelung formulation of the free-particle Schrödinger equation. A reference Born probability density and a controllable initial phase function serve as initial data for the free Madelung system, which couples the Born probability density and phase function through the Bohm quantum potential, while the phase function determines the hydrodynamic velocity field. Provided that the Born probability density remains positive and the hydrodynamic velocity field generates a unique global characteristic flow, samples drawn from the reference density and transported along the characteristic flow are distributed according to the evolving Born probability density at every time. As a consequence, randomness enters only through the initial sampling; the subsequent generation is deterministic and involves neither stochastic dynamics nor an independently parameterized time-dependent velocity field. We formulate terminal-time distribution matching as a PDE-constrained phase-identification problem and derive the underlying Hamiltonian and Fisher-information structure. For isotropic Gaussian wave packets, we obtain explicit dynamics and a necessary and sufficient condition for exact reachability of isotropic Gaussian targets by quadratic initial phase functions, together with the corresponding sampling map. For a smooth prescribed potential initial velocity field, we further establish that the characteristic flow approximates the associated first-order transport map with an O(T^2) error, both uniformly and in the 1- and 2-Wasserstein distances. A numerical Gaussian benchmark validates the fully discrete forward solver, while full-grid PDE-constrained phase identification is demonstrated for asymmetric bimodal targets.

Cyber defense in mission-critical environments requires integrated approaches capable of representing adversarial progression, defender-side uncertainty, mission impact, and defensive decision support within a unified framework. In operational domains, defenders must continuously estimate the evolving security posture while preserving the continuity and integrity of mission-critical functions under incomplete and noisy observations. This paper presents a mission-oriented cyber-defense framework for Cyber Situational Awareness (CSA) and decision support based on a three-layer integrated probabilistic model and an executable simulation prototype. The model combines: (i) an attack-graph model that represents possible adversarial progression through mission-relevant assets, (ii) an event model that transforms observed telemetry into posterior defender beliefs through Bayesian inference, and (iii) a state model that abstracts the inferred posture into conflict states and mission-risk levels. These components are connected to a one-step defensive action rule that balances estimated residual mission risk and operational cost. The framework is instantiated in a NetLogo agent-based simulation of an operational environment structured across the Tactical Edge Zone (TEZ), Mission Operations Zone (MOZ), and Enterprise Support Zone (ESZ). The proposal is assessed through mathematical consistency analysis, local robustness assessment under telemetry perturbations, and representative simulation traces. Results indicate that the framework and its implementation preserve coherent relationships between attack progression, telemetry-driven uncertainty management, mission-impact assessment, and cost-aware defensive decision support guided by mission-risk prioritization.

Open World Object Detection (OWOD) built on multimodal foundation models often suffers from semantic ambiguity caused by unidirectional text-to-vision matching, while rigid outlier penalties may over-suppress unknown objects near known-class decision boundaries. We propose CODE (Cross-Modal Calibration and Dynamic Suppression), a unified inference-time framework with three complementary components. Cross-Modal Joint Confidence Calibration injects global visual prototypes to calibrate text-driven known-class predictions. Uncertainty-Guided Universal Objectness Enhancement measures classification hesitation from local visual responses to strengthen potential unknown objects. Dynamic Outlier Suppression via Confidence Margin replaces rigid suppression with a margin-aware adjustment that preserves ambiguous out-of-distribution instances. Experiments on the Real-World Detection benchmark demonstrate that, with the OWL-ViT L/14 backbone, CODE achieves 21.7 U-mAP and 40.8 K-mAP in Task 1, surpassing the previous state of the art by 2.6 and 2.3 points, respectively.

In Abner J. Salgado and Ignacio Tomas. Diagonally implicit Runge-Kutta schemes: discrete energy-balance laws and compactness properties. J. Number. Math., 31(4):313-341, 2023, the notion of $U$-stability for Diagonally Implicit Runge-Kutta (DIRK) schemes was introduced. Here we establish the equivalence between $U$- and $B$- (algebraic) stability for two- and three-stage DIRK schemes, which then} provides suitable Bochner norm a priori estimates for $B$-stable methods. As applications, we first prove the convergence of $U$-stable DIRK schemes for linear coercive evolution problems under minimal regularity via energy estimates and compactness. Second, we apply these discretizations to gradient flows, which allow us to derive discrete local energy dissipation inequalities and provide counterexamples that demonstrate the limitations of stagewise energy monotonicity.

Vision-Language Models (VLMs) demonstrate exceptional visual reasoning capabilities, yet their inference costs escalate rapidly with the proliferation of visual tokens. Existing visual token pruning methods exhibit two fundamental limitations. First, most approaches operate exclusively post-vision encoder, leaving the substantial latency of the visual encoding phase unoptimized. Second, under strict token budgets, these methods often fail to jointly preserve holistic visual contexts and fine-grained details, leading to performance degradation. To address these bottlenecks, we propose PACE (Pixel-Adaptive Condense and Extract), a training-free inference framework that accelerates both the vision encoder and the Large Language Model (LLM) via a unified Condense-and-Extract paradigm. During the Condense stage, an Adaptive Pixel Compressor (APC) evaluates visual information density prior to encoding, adaptively downsampling redundant inputs, curtailing encoder computation while preserving global context and essential visual cues. In the Extract stage, a Dynamic Dual-Attention Extractor (DDAE) selectively retains visual tokens via a fusion of internal visual signals from the encoder and semantic signals from the LLM, safeguarding task-critical details. By integrating PACE into Qwen2.5-VL-7B, the model retains 93.8% of its original performance while utilizing only 10% of the visual tokens, yielding a 3.1x speedup in time to first token (TTFT). Our code is available at https://github.com/jjL357/PACE.

Nitrogen rates for winter wheat are set before the season, under unknown prices and weather. The standard UK advice does not respond to prices, yet recent price swings moved the most profitable rate by tens of kilograms per hectare. Machine learning is often proposed as the fix. However, it is usually judged on prediction accuracy, and accurate prediction does not by itself make the recommended rate more profitable. Our insight is to score nitrogen advice directly by the profit it forgoes on measured yield response curves. We build a test bench on 892 such curves from two long running UK experiments, and sweep the nitrogen to grain price ratio to cover all price scenarios. On this bench, machine learning fails as a predictor. No model recovers the best rate within farm tolerance, and the benchmark noise shows none can. At normal prices, every model also loses to the standard advice on profit. The gain sits elsewhere. A simple correction step applied after the model cuts profit losses by a quarter, while better models and extra features give no gain. The same frozen correction cuts losses by 43% at the second site without any retraining. A hybrid of standard advice plus a damped correction removes bias and trims rare large losses. The same price sweep also prices emission cuts, at a cost comparable to current carbon prices. Machine learning therefore pays as a profit scored correction to standard advice, not as its replacement.

A known necessary condition for Fisher consistency of the structured support vector machine requires the task loss to be a metric for which every output triple has a common geodesic point. We show that this condition is not sufficient for the canonical coordinate-wise argmax decoder. A four-output unit star admits an exactly optimal score vector whose maximizers are all strictly non-Bayes, and four outputs are minimal among metrics satisfying the condition. We then completely classify positively weighted tree metrics whose vertex set is the output space: argmax consistency holds if and only if the tree is a path. The failure on branching trees is confined to boundary distributions; every tree retains the argmax property at every full-support distribution. Among metrics satisfying the common-geodesic condition, five outputs are necessary and sufficient for a full-support counterexample; $K_{2,3}$ is the smallest member of an infinite $K_{m,n}$ family. We additionally give a full-support counterexample for the three-dimensional Hamming cube. All optimality claims have exact primal-dual certificates. The counterexamples expose a concrete decoder gap: in this polyhedral setting, an embedding can guarantee the existence of a calibrated link without validating a prescribed argmax link on every surrogate-risk minimizer.

We determine the strong converse exponent for the entanglement cost of quantum state merging, showing that it is characterized by the optimized $\alpha$-$z$ conditional Rényi entropies with $z=\alpha/2\in[1/2,1]$. This contrasts with the sandwiched conditional Rényi entropies that typically govern strong converse exponents in quantum information theory. As a consequence, we derive the strong converse exponent of the partially smoothed conditional min-entropy in purified distance. This exponent is governed by club-sandwiched conditional entropies, whereas global smoothing leads to a sandwiched expression.

Vision generative artificial intelligence (AI) has emerged as one of the most rapidly advancing areas of deep learning. The explosion of multimodal models has made them widely associated with text-to-image applications running on large datacentres. However, vision generative models are equally needed in applications that operate under strict hardware constraints at the edge, including autonomous vehicles, agricultural sensors, and mobile devices. In this Perspective, we argue that progress in vision generative AI has been driven by output quality, with hardware evolving reactively to accommodate growing model demands. We quantify the parameter cost and energy efficiency of these models across a range of accelerator platforms, and map four generative model families against seven real-world application domains. Finally, we advocate a software-hardware co-design approach, where deployment constraints are considered from the start of the design process, ensuring that the "right model" runs on the "right hardware" to serve the "right application", making generative AI deployment sustainable and accessible across a much broader range of platforms.

To achieve sustainable intelligent mobility, 6G-empowered robotic vehicles (RVs) require high-fidelity visual perception under stringent bandwidth and energy constraints. Semantic communication offers a spectral-efficient solution but suffers from severe interference in uplink non-orthogonal multiple access (NOMA) RV networks. To address this, we propose a knowledge distillation-driven and generative models-enhanced NOMA framework for robust and green RV communications, named KDG-SemNOMA. First, we develop a ConvNeXt-based deep joint source-channel coding (DeepJSCC) architecture with an enhanced attention feature (AF) module for dynamic channel adaptation. Second, to mitigate interference without inference overhead, an orthogonal transmission teacher model guides the NOMA student model via a two-stage knowledge distillation strategy. Finally, to address the over-smoothing artifacts of pixel-wise optimization, we introduce a channel-conditional GAN (cGAN). By explicitly taking the Stage-I initial reconstruction and channel states as conditional inputs, this module refines coarse outputs into high-fidelity images with realistic textures. Experiments on FFHQ-256 demonstrate that KDG-SemNOMA significantly outperforms state-of-the-art methods in both pixel-level accuracy and perceptual fidelity.

In 2025, the US government launched an unprecedented series of attacks on its own scientific research groups. A year later GitHub dropped below 90% availability for the first time, while wildfires in Canada, France, Spain, and elsewhere forced researchers from the homes and labs. These events and others have reminded us just how fragile research computing systems can be, and that planning for disasters is one of the most effective ways to prevent them. This paper is a short guide to disaster planning and recovery for a small research software team. The tips assume you are doing everything yourself on top of your regular job, and that you aren't an experienced system administrator. Some of the tips do require that kind of expertise, but most research institutions have research computing groups, data librarians, and environmental health-and-safety offices whose entire job is to help with exactly these problems. This paper tells you what "done" looks like; they can often provide it.

This paper explores design considerations for egocentric network layouts in immersive environments, providing fresh empirical insights that enhance egocentric network analysis. An egocentric network focuses on the topological and semantic relationships around a focal node (ego) and its neighboring nodes (alters), targeting local sub-networks rather than the whole network. Traditional desktop environments, limited by display constraints, often face visual clutter as node numbers grow. Building on recent findings that immersive environments enhance network analysis, we explore layouts tailored for these spaces. We begin by identifying essential design properties and dimensions for egocentric network layouts, taking into account the unique features of immersive environments. Based on these, we design four layouts-Cube, Cylindrical, Radial, and Spherical-that vary across design dimensions. We evaluate these layouts in a user study with 24 participants completing egocentric analysis tasks. Our study suggests that Cube performed well for tasks focused on ego-alter connection strength. In contrast, Spherical was more effective for understanding alter topology, minimizing occlusion, and efficiently utilizing 3D space. These findings inform design implications for future immersive egocentric network layouts.

Accurate 3D segmentation of cone-beam CT (CBCT) is critical for interventional and radiation therapy applications, yet it remains limited by two compounding challenges: the scarcity of annotated CBCT data and the large domain shift from diagnostic CT. Interventional CBCT exhibits fundamental modality differences from conventional CT, driven by acquisition and physics effects as well as contrast-specific vascular content, thereby limiting effective cross-modality model transfer. We propose a novel unsupervised domain adaptation (UDA) framework based on redundancy-reducing feature alignment, enabling 3D CBCT segmentation with no target-domain annotations or inference-time adaptation. Our framework is architecture-agnostic, seamlessly adapting both CNN-based and ViT-based foundation models. We evaluate our method on two challenging CT-CBCT liver segmentation benchmarks: one for interventional vascular procedures and one for radiation therapy, demonstrating that even large-scale pretrained segmentation networks require explicit feature-space bridging to generalize across acquisition modalities, and that our approach consistently outperforms existing pretrained foundation model and UDA strategies. To support reproducibility and benchmarking, we release the liver segmentations for a public CBCT dataset, along with the code, trained models, and weights.

Peer effects are difficult to estimate when interaction graphs evolve because pre-assignment network history, dynamic peer exposure, and post-assignment network change have distinct causal roles. We introduce a controlled contrast framework that indexes potential outcomes by own treatment, temporally aggregated peer exposure, and a post-assignment evolution summary. Differences between the resulting means define own-treatment, peer-exposure, controlled network-evolution, and joint controlled contrasts rather than a mediation decomposition. We develop the Dynamic Network Doubly Robust estimator, DynaNet-DR, which combines a temporally factorized propensity with normalized augmentation. Under consistency, summary sufficiency, sequential exchangeability, positivity, nuisance convergence, and weak dependence, its canonical estimator is consistent when either the outcome regression or the propensity estimator is consistent. The reported implementation adds representative-score prediction, fixed clipping, and finite-sample stabilization. Semi-synthetic benchmarks on fixed real temporal graph sequences show favorable estimation accuracy among methods targeting the full profile. These benchmarks assess summary-indexed contrasts rather than counterfactual edge generation, and the MathOverflow study is an observational illustration under the stated assumptions.

Soft actuators enable dexterous and compliant interaction, but closed-loop task-space control remains challenging due to strong nonlinearities, distributed deformation, and uncertainty in their dynamics. This paper presents a real-time dynamic-model-based task-space feedback and estimation framework based on a non-minimal coordinate discrete elastic rod model formulated in absolute coordinates with holonomic constraints. The resulting structure preserves distributed mechanics while maintaining computational efficiency through sparse system matrices, enabling real-time control with up to 10 discretized rods. A quasi-static feedforward inverse model is combined with a task-space PI controller and a dynamic observer that fuses measurement residuals as virtual forces, enabling full-state estimation from sparse sensing. The approach is experimentally validated on three planar pneumatic soft actuators with varying geometries. Across five tasks, including drawing the digits 0-9 across the workspace (3-18 mm/s tip speed), tracking periodic motion (up to 37 cm/s), cross-platform generalization, reduced sensing conditions, and real-time user-defined references, our method achieves 1.5-2.3 mm root mean square error (RMSE) for precision motions and 5.5-12.4 mm RMSE at 1-2 Hz. Results demonstrate that structured, non-minimal dynamic models can enable real-time, high-precision, moderate-bandwidth task-space control of planar soft pneumatic actuators in free space.

In closed cattle-fattening barns, the indoor climate and herd growth are mutually interdependent. Temperature, relative humidity, airflow, and ventilation affect thermal comfort, feed intake, metabolic heat production, daily growth, feed efficiency, and energy consumption, while body-weight gain alters the future heat and moisture loads of the barn and, consequently, its ventilation, heating, and energy requirements. This article proposes a trans-domain digital twin framework with single-episode learning capability, customized for bio-aware climate and energy control in a closed cattle-fattening barn. The framework integrates a mechanistic climate simulator, a livestock growth simulator, model predictive control, lightweight reinforcement learning, and structured knowledge memory within a multi-rate temporal-loop architecture. The fast temporal loop operates every five minutes to evaluate actuator decisions and maintain short-term thermal comfort, safety, and energy efficiency, whereas the slow temporal loop provides biological guidance based on daily climatic conditions, feed efficiency, heat production, and growth-limiting factors. The results show that climate, growth, energy, feed, biological guidance, and memory can be linked within a single executable control cycle. Remaining limitations include the need for field validation, improved management of feed pressure, and reduction of abrupt actuator-command variations.

Electronic design automation (EDA) has advanced engineering productivity through successive generations of tooling that progressively automate synthesis, optimisation, and verification. Large language models (LLMs) extend this trajectory by enabling direct translation from design intent to hardware implementations. In most of the EDA literature, LLM-based solutions are typically assisting siloed design stages or tasks, however this obscured the drivers by which capability emerges and systems scale. In this Perspective, we instead define three hierarchical roles that reveal how capability accumulates: a Generator that produces design artifacts in a single pass, an Agent that refines outputs through iterative tool feedback, and an Orchestrator that coordinates decisions across EDA-stages. Across published systems, this reveals a syntax trap in which models are trained to produce plausible code rather than physically correct hardware, compounded by fragmented tools and loss of design context that obscure how decisions affect later stages. Comparisons across the three roles show that current approaches struggle to scale to industrial designs, motivating a shift towards a standardised, physics-aware orchestrator that connects tools and agents across the EDA flow for more reliable and accessible hardware design.

LLM agents are increasingly applied to anomaly detection and root-cause analysis in time-series observations collected from real-world systems; however, their performance on these tasks has not been systematically evaluated under controlled conditions. We introduce TraceBench, a simulation-based framework for generating controlled root-cause attribution tasks. In each generated task, an agent receives time-series observations produced by simulating a physical dynamical system and must determine whether a system parameter was altered during the simulation and, if so, which one. Using TraceBench, we generate tasks from three interpretable mechanical systems and systematically evaluate four LLM agents across controlled experimental conditions, yielding new insights into how these agents analyze time-series observations from dynamical systems. Our results show that agents benefit substantially from domain context and explore data primarily through numerical console output rather than visualizations. We also find that agents generally perform worse when required to produce a Python script that maps each time-series sample to a predicted root-cause label than when they submit predictions directly. We release our datasets, agent trajectories, experimental results, and a leaderboard on our website, tracebench.github.io.

Keypoint detection under motion blur remains a significant challenge, as blur distorts local image structure and degrades the repeatability of feature localization. Existing approaches either rely on computationally expensive deblur-then-detect pipelines that may introduce restoration artifacts, or learn to regress the image positions of handcrafted keypoints extracted on sharp images, which reflects the assumptions of the handcrafted detector rather than what is truly repeatable under blur. We present SSMB, a deblur-free, self-supervised keypoint detector for motion-blurred images that requires neither handcrafted detectors nor external pseudo-labels. SSMB introduces the Local Discriminability Enhancement (LDE) module, which restores fine-grained local discriminability after global feature mixing. Training is performed in two stages. First, geometric pretraining on synthetic shapes bootstraps spatially discriminative keypoint detection without any external detector, just from the rendered geometry. Second, blur-aware training on real sharp-blur image pairs learns blur-invariant detection through a multi-component self-supervised objective that enforces cross-domain consistency, geometric alignment, and spatial coverage. Extensive evaluations on keypoint detection, image matching, relative pose estimation, and visual localization under motion blur demonstrate that SSMB establishes a new state-of-the-art among sparse keypoint detectors, consistently outperforming both supervised and self-supervised baselines across all tasks. Code, models, and datasets will be publicly available upon paper acceptance.

A functor is a familiar model of an implementation, where every operation in a specification is assigned a concrete instantiation at the outset. But some tasks are less eager: we often want to assemble systems while leaving backend choices such as data representations and algorithms for later. We observe that cofunctors admit a direct reading as such partial implementations, which are implementations whose extra argument is a state-dependent family of deferred choices. While cofunctors (also called retrofunctors) are not novel, their conceptual and purposive reading in this manner appears to be.

In this paper, we resolve a 30-year-old conjecture of Spielman and Teng concerning the performance of the spectral partitioning method on graphs embeddable on an orientable surface of genus $g\ge 1$. In particular, for such a graph $G$ with $n$ vertices and maximum degree $\Delta$, we show that the second-smallest eigenvalue of its Laplacian matrix satisfies $\lambda_2(L_G)\lesssim\Delta\frac g n$. We also obtain an improved eigenvalue bound for $K_h$-minor-free graphs of $\lambda_2(L_G)\lesssim\Delta\frac{h^2(\log h)^2}n$. In fact, our results directly prove much stronger results for reweighted eigenvalues, including higher reweighted eigenvalues. As a consequence, we obtain bounds not just on Laplacian eigenvalues, but also on normalized Laplacian eigenvalues and Steklov eigenvalues. Our results for genus-$g$ graphs are optimal for all of these kinds of eigenvalues, while our results for $K_h$-minor-free graphs are optimal up to $\log(h)$ factors. Our techniques for genus-$g$ graphs bootstrap bounded-degree bounds of normalized eigenvalues for entire classes to bounds for reweighted eigenvalues for the same classes without the bounded-degree limitation, while our techniques for $K_h$-minor-free graphs generalize an argument of Korhonen and Lokshtanov, making use of the Lovász local lemma.

Accurate rectal cancer segmentation from magnetic resonance imaging (MRI) is essential for adaptive radiotherapy and tumor response assessment, but deployment also requires computational efficiency and informative, calibrated uncertainty estimates. We therefore introduce SWIFT, a SWin pretrained model wIth parameter-eFficient and Tumor-aware fine-tuning for rectal cancer segmentation. A Swin V2 encoder pretrained on 10,444 public 3D CT volumes using a DINOv2-style objective was adapted to T2-weighted MRI through four cumulative configurations: full fine-tuning (SWIFT), decoder compression (SWIFTe), low-rank adaptation (SWIFTe-LoRA), and a four-member LoRA-decoder ensemble (SWIFTe-LDE4). Geometric accuracy, tumor detection, radiomic agreement, and probability calibration were evaluated on a held-out 247-case test set from a single-institution cohort acquired using 1.5 or 3 Tesla GE scanners. Compared with SWIFT, SWIFTe reduced total parameters by 70.1% (from 72.8M to 21.8M) and increased tumor detection rate from 89.9% to 93.9%, while achieving a slightly lower median surface DSC (0.61 versus 0.62) and improved radiomic agreement. In a separate SWIFTe ablation, removing tumor-aware augmentation reduced detection from 93.9% to 89.9% but increased surface DSC from 0.61 to 0.64, demonstrating a detection-boundary-agreement trade-off. SWIFTe-LoRA used 14.6% of SWIFTe's trainable parameters while retaining similar segmentation performance. SWIFTe-LDE4 achieved the lowest calibration errors among the four configurations after temperature scaling (expected calibration error, 0.217; Brier score, 0.222), although the absolute expected calibration error indicates residual miscalibration. Similar efficiency-calibration patterns were observed using the public VoCo checkpoint, supporting robustness across pretrained initializations rather than external clinical generalizability.

Understanding spoken dialogue requires joint reasoning over lexical content and paralinguistic acoustic signals such as emotion and conversational intent. However, existing evaluations often allow shortcuts based on transcripts or single-modality solutions, obscuring whether models genuinely ground predictions in speech. We formalize this failure mode as cross-modal disagreement, where transcripts suggest plausible but incorrect surface interpretations while acoustic cues such as prosody or speaking style support different answers. We develop a scalable framework that identifies text-biased surface interpretations and converts disagreement regions into conflict QA examples. We also include consistent cases where transcript-based and speech-grounded interpretations agree, enabling evaluation beyond adversarial audio dependence. This results in ContraTalk, a controlled benchmark containing 501 questions across five discourse dimensions: interaction behavior, emotion state, dialogue act, social stance, and conversational intent. We further develop an agentic-style reasoning framework that converts speech into an Audio Twin, a text-readable representation of localized acoustic cues that exposes acoustic evidence to the reasoning model. Experiments show that strong text-only LLMs exceed 90% accuracy in consistent cases but drop to 33-48% in conflict cases. Direct AudioLLMs provide only partial grounding, still selecting the transcript-biased trap in roughly 30-40% of conflict cases. Our Audio Twin framework improves conflict-case accuracy while reducing trap selection, but its consistent-case behavior remains backbone-dependent. These results identify transcript-based shortcuts as an important failure mode in spoken dialogue understanding and show that explicit acoustic evidence aggregation provides a more controllable interface for diagnosing and improving speech-grounded reasoning.

Many Earth Observation applications need land-use/land-cover maps that are both precise and frequently updated, yet the strongest Earth Observation foundation models build their embeddings from a full year of observations. We present a controlled study of the temporal sensitivity of Tessera, one of these leading foundation models, for land-use/land-cover mapping. Keeping the encoder frozen, we recompute its embeddings over varying observation windows, from a full year down to a single day. We use them as inputs to a linear probe and a UNet segmentation head, benchmarking both of them against from-scratch networks on LUCAS, DynamicEarthNet, and PASTIS-R datasets. We show that the value of the embeddings is task-dependent. Where classes are separated by phenology, as for the crop types of PASTIS-R, they reach a mean Intersection-over-Union of $58.3$, about $46\%$ above the best from-scratch model. Where classes are temporally stable (e.g., forests in DynamicEarthNet and LUCAS), embedding-based and from-scratch models match only under full supervision. On both datasets, Tessera embeddings remain markedly more label-efficient. Degradation under shorter temporal windows is gradual and class-dependent. Contracting the window from one year to one month costs $39\%$ of the segmentation accuracy on PASTIS-R but only $5\%$ on DynamicEarthNet. Single-day embeddings still classify land cover in LUCAS at $3.4$ times the chance level. Our study shows that temporal coverage is therefore a tunable cost rather than a fixed prerequisite, opening regimes such as near-real-time mapping and faster land-use/land-cover refresh cycles.