Papers

Vision foundation models trained on Internet-scale RGB datasets enable remarkable capabilities across a range of tasks, from text-to-video generation to few-shot 3D scene reconstruction. An acoustic foundation model trained on large-scale sonar datasets could enable similar capabilities in the underwater domain, where turbidity and low-visibility conditions make conventional RGB foundation models inapplicable. Unfortunately, a lack of freely available large-scale sonar datasets makes training such a model from scratch impractical. In this work, we demonstrate that vision foundation models can be efficiently adapted to the sonar setting by (1) exploiting the geometric relationship between the two sensing modalities and (2) employing accurate physics-based noise models for synthetic data generation. The resulting sonar adaptation models enable new capabilities: For the first time, we experimentally demonstrate sonar-based pose-free 3D reconstruction.

Reliable marine-engine fault diagnosis in maritime IoT is challenged by distributed data ownership, heterogeneous fault distributions, and continuously changing operating conditions. This paper proposes SeaCausal-FL, a federated fuzzy causal learning framework that combines a shared temporal diagnostic path with mechanism-conditioned causal reasoning. An interval type-2 fuzzy layer represents uncertain and overlapping operating mechanisms, while each mechanism is associated with a physics-constrained structural causal model. Before aggregation, locally learned mechanisms are aligned using operating context, causal structure, and conditional intervention-response signatures. Model parameters are then aggregated according to sample, class, mechanism, and mechanism-class evidence instead of client sample size alone. The learned structural equations further support interval counterfactual reasoning through abduction, action, and prediction. Experiments on a marine-engine fault dataset and a real-data-calibrated semi-synthetic causal benchmark show that SeaCausal-FL achieves an average F1 score of 87.07% across four client partitions, with AUROC and AUPRC of 98.98% and 94.81%, respectively. It also maintains strong performance under unseen loads and fault-type omission during training. On the causal benchmark, SeaCausal-FL reaches an Edge-F1 of approximately 0.58 and an Edge-AUPRC of 0.68, reduces coefficient RMSE to about 0.14, and provides favorable counterfactual estimation and intervention decisions.

Vision-language-action (VLA) models have shown promising generalization for language-conditioned robot manipulation, but deploying them in unstructured environments remains challenging. A single end-to-end VLA policy must simultaneously solve long-range transport of the end effector to task-relevant regions and short-horizon, contact-rich interaction upon arrival. This formulation is inefficient and brittle: small visual shifts, distractors, clutter, occlusions, or unfavorable initial gripper poses can push the policy outside the local state distribution in which it was trained, leading to task failure. We introduce GloVLA, a hybrid framework that explicitly separates object-centric manipulation into two complementary regimes: a geometric transport controller moves the end-effector into interaction-centric handoff regions, and local VLA policies handle only the short-horizon interaction phases. GloVLA is model-agnostic and can be integrated with different VLA backbones with no additional demonstrations and no changes to the action space or success predicate. Experiments on standard LIBERO and LIBERO-Plus Object tasks together with a newly introduced LIBERO-Challenge benchmark ettings with clutter, distractors,illumination changes, visual shifts, and obstruction show that GloVLA improves task success and substantially lowers VLA inference cost compared with full end-to-Challenge, full-trajectory GR00T N1.6execution degrades to 20.9% average success while GloVLA retains 88.5%; on a physical UR10e, overall success improves from 35.6% to 90.0% while mean inference time is more than halved. Videos and additional results are available at https://glovla-project.github.io/

Translating between imaging sonar and optical cameras is valuable for underwater perception, but supporting both directions with separate models duplicates storage and computation. A unified bidirectional model is therefore attractive, yet existing approaches largely treat the two directions symmetrically despite their fundamentally different image-formation physics. We argue that sharing a generative model does not require sharing the physics. We introduce the Direction-Asymmetric Realism Bridge (DARB), which retains a shared diffusion-bridge trunk while routing direction-specific physical priors through asymmetric pathways: range-aware modulation for sonar-to-optical translation and polar ray-dependent processing for optical-to-sonar translation. We further show that symmetry in training is also costly: applying a common realism schedule reduces sonar-to-optical PSNR by 2.60 dB. Our Adaptive Realism Supervision (ARS) instead determines when, where, and how strongly perceptual supervision is applied from reconstruction quality and gradient balance. Together, DARB and ARS enable one bidirectional model to match the sonar-to-optical specialist within 0.11 dB PSNR, outperform the optical-to-sonar specialist by 0.70 FID, and surpass two independently trained BBDMs on seven of eight metrics.

Grounding natural-language instructions into reliable and executable actions remains a fundamental challenge for vision-language-action (VLA) systems on mobile robots, due to the persistent gap between high-level semantic reasoning and low-level locomotion and manipulation control. Existing approaches often rely on implicit reasoning or monolithic action prediction, making it difficult to maintain coherent long-horizon decision making while producing precise and adaptable robot actions. To address this challenge, we propose MobileVLA-R1 2.0, an RL-enhanced VLA framework that explicitly couples structured embodied reasoning with executable mobile robot control. The framework learns multi-granularity reasoning over embodied trajectories through supervised Chain-of-Thought (CoT) alignment and reinforcement learning, improving reasoning-to-action consistency beyond purely behavioral supervision. To support both locomotion and manipulation, we further introduce a reasoning-conditioned action decoder that maps multimodal reasoning representations to task-level action targets, which are subsequently translated into embodiment-specific commands by robot controllers. This design provides a unified perception-reasoning-action interface while decoupling high-level action generation from robot-specific actuation. We conduct extensive evaluations on language-guided navigation, quadruped control, and humanoid mobile manipulation, covering VLN-CE, QUARD, and real-world deployments on Unitree Go2 and G1 robots. MobileVLA-R1 2.0 consistently outperforms strong VLA baselines, achieving an average 1.6 point improvement in SR on VLN-CE and a 10.0 point improvement in full-task success on real-world G1 mobile manipulation tasks over MobileVLA-R1, while demonstrating robust long-horizon instruction following and closed-loop execution across different robotic platforms.

Security Operations Centers (SOCs) process large amounts of tickets, most of which are low-interest events not worthy of further investigation. The repetitive nature of this task and similarity of the vast amounts of tickets make it a prime candidate for generative AI-based automation. We created and deployed an agentic AI companion utilizing large language models through fieldwork within a SOC for over one year. The design of the SOC AI companion was driven by researchers' participation and interactions within the SOC's daily work. SOC analysts were invited to use it during the last four months of the fieldwork. We analyzed the analysts' usage of the companion and found that in more than 90% of the cases the companion's outputs were reused by analysts in the ticket's closing report. Our results showed that when designed "in the trenches" with the intended users, a SOC AI companion can go beyond being yet another tool, but rather a system that co-evolves with its human users as it traverses through the various types of workloads. Analysts naturally started to shape the AI companion's behaviors to fit their particular needs. Our data show that the more human analysts shape the AI companion's behaviors, the more they become comfortable trusting the output from the AI system, resulting in improved productivity.

We numerically implement full quantum circuits of optimal tapered Quantum Phase Estimation (in the form of an implementable approximate bandwidth-limited DPSS taper) for the purpose of minimum eigenvalue computation, where the Hamiltonian time-evolution unitary is implemented using standard Trotterization, and the taper state is constructed using an optimized bandwidth-limited taper. We illustrate, on representative quantum Hamiltonians, with tractable exact classical numerical quantum circuit simulations, how tapered QPE performs with respect to optimal phase sampling rate, absolute error, Shannon entropy of the optimal phase distribution, when the tQPE algorithm parameters are changed. Those algorithm parameters are Trotter error, finite sampling, imperfect initial states, total evolution time, and digitization error (number of phase qubits). The reported numerical experiments include up to $\ell=10$ phase qubits of precision with $m=3$ additional phase-register error suppression qubits, on a $4$-qubit Heisenberg quantum magnet model Hamiltonian, using up to 12th-order Trotterization. We numerically show i) the steady-state optimal phase sampling rate is determined by the initial state overlap with the ground-state, ii) with respect to phase sampling probability, or overall eigenvalue sampling error rate, there is no strong evolution time dependence for tQPE unless there is high Trotter error, iii) the approximated time-evolution unitaries in tQPE, like in standard QPE, can result in substantial ringing, which leads to non-physical eigenvalue estimates.

Multimodal Large Language Models (MLLMs) perform strongly on general visual understanding tasks such as visual question answering, yet they often struggle with a basic comparative skill: identifying what has changed between two similar images. We introduce VDiff-Bench, a challenging multiple-choice benchmark for fine-grained Image Difference Identification. VDiff-Bench contains 1,756 four-way questions over image pairs and covers 10 change categories: position, motion, regional image color, overall image color, appearance/disappearance, noise/resolution, texture, substitution/size, OCR/text, and illumination. Each question corresponds to two image inputs with 4 choices: the true difference, two hard negative descriptions, and a "no difference" distractor. To make the task challenging, we specifically curate ground-truth-conditioned negatives that require models to distinguish the actual change from nearby semantic alternatives. Experiments with 11 state-of-the-art open- and closed-source MLLMs show that fine-grained visual comparison remains brittle: models exhibit uneven performance across sources and change categories, with persistent failures on subtle low-level changes like noises and textures. For instance, three 7-8B-scale open-source MLLMs score 52.5-70.6% on semantic changes but only 8.7-33.3% on low-level changes like noise and texture, falsely assuming no changes between two image inputs. Surprisingly, despite strong performance of other closed-source commercial models, Grok 4.3 demonstrate remarkable performance drop on identifying noise and texture differences between images, falling significantly behind large open-source models like Kimi K2.5 and K3. Overall, VDiff-Bench provides a targeted diagnostic for evaluating comparative visual understanding in MLLMs, exposing failures that are not captured by standard single-image vision-language tasks.

The $\mathbb{Z}_2\mathbb{Z}_4\mathbb{Z}_8$-additive codes are subgroups of $\mathbb{Z}_2^{\alpha_1}\times\mathbb{Z}_4^{\alpha_2}\times\mathbb{Z}_8^{\alpha_3}$. A $\mathbb{Z}_2\mathbb{Z}_4\mathbb{Z}_8$-linear Hadamard code is a Hadamard code which is the Gray map image of such a code. A recursive construction of $\mathbb{Z}_2\mathbb{Z}_4\mathbb{Z}_8$-additive Hadamard codes $\mathcal H^{t_1,t_2,t_3}$, with all $\alpha_i\neq0$, $t_1\geq1$, $t_2\geq0$, and $t_3\geq1$, is known, as are the linearity and kernel of the corresponding codes $H^{t_1,t_2,t_3}$ of length $2^t$. Their rank, however, was not known in closed form. In this paper, we determine $rank(H^{t_1,t_2,t_3})$ for every admissible triple and obtain an explicit closed formula. The proof computes the dimension of the space spanned by the binary coordinate functions of the code, viewed as Boolean functions of the binary digits of the message. We then use the rank formula, together with the known kernel dimension, to classify the family as far as these two invariants allow. The only pairs of distinct types of the same length sharing both invariants are $\bigl(H^{1,2,t-6},H^{2,0,t-5}\bigr)$ for $t\geq7$ and $\bigl(H^{2,2,t-9},H^{3,0,t-8}\bigr)$ for $t\geq10$. Consequently, the number of distinct pairs $(r,k)$ among the codes of length $2^t$ is $\lfloor(t^2+6)/12\rfloor$ for $3\leq t\leq6$, $\lfloor(t^2+6)/12\rfloor-1$ for $7\leq t\leq9$, and $\lfloor(t^2+6)/12\rfloor-2$ for $t\geq10$. Thus, rank and kernel dimension classify the family completely if and only if $3\leq t\leq6$. We also determine all coincidences with the $\mathbb{Z}_4$-, $\mathbb{Z}_2\mathbb{Z}_4$-, and $\mathbb{Z}_8$-linear Hadamard codes of the same length: there is exactly one infinite family with the first two, namely $H^{2,1,t-7}$ and $H^{5,t-9}$ for $t\geq10$, and exactly five families with the $\mathbb{Z}_8$-linear codes.

We introduce the weighted Laplace space $H_w$, an RKHS of Laplace transforms on $(0,\infty)$, and study spectral measures in its dual space $H'_w$. For conforming FEM discretizations of the Dirichlet Laplacian on bounded Lipschitz domains, we prove the dual-norm inequality $\|\mu_h\|_{H'_w} \leq \|\mu\|_{H'_w}$, where $\mu = \sum_k\delta_{\lambda_k}$ and $\mu_h = \sum_k \delta_{\lambda_{k,h}}$. The proof combines min-max monotonicity of FEM eigenvalues with a heat-trace representation of the dual norm. We then analyze $H_w$-adapted rational approximation of shifted symbols $\phi(x)=(x+\kappa^2)^{-\beta}$ and give a conditional transfer principle for estimates proved in the corresponding weighted Laplace pre-image norm. Via dual pairing, the norm inequality yields uniform bounds for finite spectral sums and related transformed observables.

Connected and automated vehicle platoons require reliable dissemination of task-relevant hazard information to enable appropriate downstream responses, with unresolved situations additionally requiring remote human intervention. However, the limited channel bandwidth of V2X communication makes high-volume sensor data transmission challenging, motivating the exchange of only task-relevant information rather than raw sensor data. This paper proposes an integrated semantic and task-oriented communication framework for cooperative hazard response and remote operation in automated vehicle platoons. The framework combines onboard instance segmentation and monocular depth estimation for task-oriented hazard decision-making with Stable Diffusion Variational Autoencoder (SD-VAE)-based semantic scene compression. Task-relevant hazard information and the compact semantic representation are jointly disseminated through a broadcast-based Emergency Semantic Message (ESM) over the existing V2X protocol stack. Evaluation demonstrates that SD-VAE consistently outperforms JPEG and JPEG 2000 on task-relevant reconstruction metrics, exhibiting graceful degradation rather than a sharp cliff effect under low-SINR conditions, demonstrating its feasibility for remote scene reconstruction when human intervention is required. Moreover, an extensive campaign of 7,290 simulation runs shows a low overall inter-vehicle collision rate of 0.99%, with collisions occurring mainly under extreme high-speed and short-gap conditions.

We introduce LatentGamma, a differentiable operator on the unit cube $[0,1]^M$ designed as a smooth surrogate of Tarski's immediate consequence operator $\TP$ associated with a definite Horn program $P$ on $M$ atoms. The operator is built as a five-stage composition combining sigmoidal gating, softmax routing, and a residual update that enforces monotonicity by construction. We establish four theoretical results. First, the iterated sequence is coordinate-wise non-decreasing and bounded by $\ind$, hence converges to a fixed point of LatentGamma; the operator itself is lattice-monotone on $[0,1]^M$. Second, our constructive expressivity theorem shows that for every definite Horn program $P$ there exists a closed-form parameter assignment $\theta^*(P)$ and an explicit time bound $T_{\max}(P)$ such that the iterated sequence exactly reproduces $\TPinf(F_0)$ for every initial fact set $F_0$ throughout the window $[D(P, F_0), T_{\max}(P)]$, where $D(P, F_0) \leq M$ is the derivation depth. The window $T_{\max}$ is large in practice ($\geq 10^5$ for sparse programs) and reflects the finite-time nature of computation by smooth sigmoidal gates. Third, the oracle is robust to Gaussian noise on its body and head logits, with explicit non-asymptotic bounds. Fourth, we prove a matching information-theoretic lower bound on the parameter count. We provide a complete numerical validation on programs ranging from $M = 33$ to $M = 504$ atoms: oracle accuracy reaches $1.0000$ on $2000$ test cases with zero false-positive and false-negative rates, and the empirical noise tolerance $\sigma_{\max}$ scales precisely as the union bound $RM \cdot \Phi(-10/\sigma_\eta)$ predicts.

Ground-truth defect masks in industrial inspection datasets are typically reserved for evaluation. This paper repurposes them as spatial supervision signals during training of classification networks, teaching a model not just what to predict but where to look. The method adds an activation-based attention alignment loss that steers convolutional feature maps toward defect regions, in a mixed-supervision formulation that also accommodates samples without masks, such as diffusion-generated images. Combined with DDPM augmentation, synthetic images contribute quantity while masks contribute spatial precision. We evaluate 85 models (four CNN backbones under a 2x2 data/training factorial over five seeds, plus a Swin-V2-T transformer baseline) on the MVTec-AD bottle benchmark, with localization measured on held-out defect images excluded from classifier gradient updates. Main findings: (1) attention-guided training improves activation-based localization (Pixel-AUROC) by +18.0% for EfficientNetB0 with augmentation (p=0.005, Cohen's d=2.6) and +18.7% for ResNet50 (p=0.008), significant in four of eight CNN settings (uncorrected for multiple comparisons) with no significant change in classification; (2) for EfficientNetB0 a data x training-mode interaction is significant (p=0.002), consistent with a super-additive effect (+13.6% combined vs +1.6% summed individual effects); (3) architectures with weaker spatial representations benefit most, whereas ConvNeXt-T shows no effect, apparently because its depthwise-convolution activations yield spatially uninformative channel-mean maps; (4) unsupervised PatchCore remains the strongest localizer (Pixel-AUROC=0.983), contextualizing the supervised gains. These results show that existing evaluation masks can act as practical training signals that measurably and reproducibly improve where defect classifiers attend.

Two centuries ago Pierre-Joseph Proudhon envisioned a social contract that is (1) an agreement of man with man; (2) reciprocal; (3) imposing no obligation upon the parties except that which results from their personal promise; (4) subject to no external authority; (5) freely accepted and signed by all the participants; (6) of the nature of a contract of exchange. He also envisioned a property of social contracts, analogous to a property of digital platforms we term grassroots: that one could "make a contract with all, as ... with some"; digitally, that a grassroots platform can have multiple instances, which may coalesce by mutual consent. We define grassroots social contracts as social contracts meeting these conditions and that (7) people are free to deal with each other; and (8) there is no external register of people. We show that a grassroots social contract can be transformed into a working grassroots platform through an abstraction cascade, from the contract text to formal act schemas, verified syntactically to be grassroots, and then to volition-guarded multiagent atomic transactions. Act schemas are a formal language for the acts the contract describes, each naming the parties' roles, which of them must will the act, and its precondition and effect at each role. Any contract written in this language meets the eight conditions, provided it is syntactically grassroots, satisfying three decidable conditions, Introduction, Provenance and Volition. We prove that the protocol realising a syntactically grassroots contract is grassroots, and volitionally so. Clauses of a grassroots social contract are of three legal types: a breach of an enforced clause is impossible with a correct implementation; a breach of an attested or undertaken clause can be taken to court, with signed evidence for an attested clause. We illustrate them with a grassroots social graph and grassroots currencies.

Vulnerability discovery is becoming an important ability of large language model (LLM) agents: agents that silently miss real defects leave critical software exposed. Rigorously measuring this ability is therefore urgent, but existing benchmarks are gameable through data contamination, score recall against an unknowable vulnerability set, often rely on synthetic bugs, and report a single end-to-end verdict that cannot localize where an agent fails. Vulnerability discovery is a composite ability: an agent must comprehend source code, infer input constraints, construct inputs, execute them, and iteratively correct from feedback. We recast its measurement as an input-prediction task with a closed, deterministic ground truth: using coverage-guided fuzzing, we mine deep target branches in real-world C/C++ programs and ask an agent to predict an input that drives execution to a given branch. This decomposes discovery into three task modes over 22 real-world C/C++ programs spanning 15 domains. Open-loop and Feedback-enabled share 60 fixed-target task instances across 16 of these codebases (13 domains), testing input construction without and with a distance oracle to isolate code comprehension from feedback-driven correction. Online Arena instead removes the predefined target and scores path exploration by coverage gain on a separate, partially overlapping pool of 11 programs; agents collectively confirmed 13 distinct bugs across six programs. Evaluating 15 default-effort model-scaffold configurations, the best reaches only 55.0% pass rate in the Feedback-enabled mode, and the mean across seven paired Claude Code configurations is 36.4% with feedback versus 19.3% without. Decomposing failures, we find constraint inference, not navigation, is the dominant bottleneck. We release SWE-Test with a turnkey evaluation environment.

Large language models (LLMs) have recently made substantial progress in formal proof generation, yet presenting distinctive challenges in cryptographic area. Computational security arguments posit that a valid proof must coordinate probability, adversarial games, invariants, assumptions and bounds, which can be provided by a machine-checked framework named EasyCrypt. Although all objects may appear in available context, LLMs still struggle because proof-theoretic dependencies are typically implicit in a linear representation and distributed across multiple programs. So, this paper presents Scratchy, a visual-scratchpad approach that exposes these dependencies for multimodal generation. Given the natural-language security description, with formal context and target propositions, the proof objects can be normalized into a typed proof-relation graph. Then a structure-preserving visual compiler transforms the graph into the formula-rich visual proof state that guides a multimodal model in generating the EasyCrypt proof. Also, the Scratchy-eval, a 114-task dataset derived from reliable official EasyCrypt files, has been introduced. It contains 64 security-form proof generations and 50 multiple-choice knowledge tests. After a series of evaluations, covering semantic grounding, relational invariants, and game reductions, classical LLMs like GPT-5.6-Sol and Claude-Opus-5 have gained a clear advantage from Scratchy's structured visual proof states. This contrast suggests that explicit proof structure can make the improvement and multimodal proof-state representation as a promising direction for computer-aided cryptography.

AI systems increasingly answer scientific questions about health, safety, and the environment. But most retrieval-augmented generation systems are tuned to provide factually correct, on-topic answers rather than to help non-experts understand what those answers mean for their lives and decisions. We focus on consequential scientific questions whose results directly shape people's lives and study them through a public water-quality communication system, where residents and community leaders interpret the findings and choose actions. Their experiences show that on-topic answers can still be insufficient without explanation and context and that the emotional weight of risk information cannot be ignored. Our system first classifies each question by reasoning type (for example, causal versus policy-based), then generates follow-up questions to identify missing evidence and retrieve it. One component clearly distinguishes between what is known and what is uncertain, while another rewrites scientific details into accessible language, using persona-based styles, such as a caring neighbor or an administrative official, to adapt tone and readability. Ablations on over $160$ questions show that the system uses $\textit{an order of magnitude less context}$ and, in several configurations, improves human-rated completeness. A completeness metric co-designed with community members and a fine-tuned learned judge reveal that standard relevance scores explain about $1\%$ of variation in human completeness ratings, and even the tuned judge only moderately aligns with humans, indicating that completeness is a distinct human-centered objective that current metrics do not reliably capture.

Vision-language-action (VLA) models have substantially advanced language-guided robot manipulation, yet reliable execution still hinges on identifying which physical object an instruction refers to. In cluttered scenes containing repeated objects, ambiguous anchors, or frame-dependent spatial terms, a robot can execute a geometrically valid action on a semantically compatible but unintended instance; we call this failure an identity switch. The referent is jointly determined by three coupled latent variables: the target, the anchor, and the reference frame, so committing to any one of them before execution turns residual ambiguity into a silent and irreversible error. We propose RefGuard, an identity-aware grounding framework that delays commitment by maintaining a joint posterior over all three variables. RefGuard builds a frame-conditioned object-centric scene graph from RGB-D observations, separating frame-independent geometry from directional relations, and routes the posterior through a decision policy that executes, clarifies, reobserves, or aborts. On a real UF850 arm, RefGuard records no identity switch on any ambiguity-stress trial and executes correctly on 90.0% of them, whereas fine-tuned VLA and LLM (Large Language Model)-based baselines switch identity in 33-46% of the same trials, while retaining 93.3% success on unambiguous scenes and recovering from post-grounding scene changes in 86.7% of trials. On a 3200-episode procedural suite, it raises correct execution on solvable instructions from 56.6% to 80.5% over the ablation that commits to the anchor and frame before the target, while deferring less often (19.5% vs. 43.4%).

Sequential recommenders are typically trained on long user histories to capture rich behavioral signals, yet serving with training-length sequences is often impractical due to real-time efficiency constraints. Directly using only recent behaviors leads to a severe performance drop. To bridge this gap, existing approaches compress user histories into persistent per-user states, storing and retrieving them at inference time; while effective, they impose non-trivial infrastructure overhead and offer little remedy in cold-start scenarios. In this paper, we empirically identify two structural flaws rooted in geometric properties and dataset sparsity, and propose a novel two-stage framework to close the long-short-view performance gap. Specifically, in the first stage, we replace the commonly used dot-product with angular similarity scoring and leverage a modified softmax to counter prefix position bias. In the second stage, we fine-tune only bias and LayerNorm components, which are universal to standard sequential backbones, for further improvement. Both stages are guided by carefully designed learning objectives. Extensive experiments on two representative backbones across three public datasets demonstrate the effectiveness of our proposed framework.

Automatic skeleton generation involves predicting both joint positions and skeletal connectivity. However, existing approaches struggle to encode branch structures into token sequences and do not use test-time computation effectively. We study these choices within a unified autoregressive framework. First, we introduce branch-centric tokenization, a branch-aware representation that places structurally related elements next to each other and encodes connectivity directly in the sequence. Compared with standard BFS-style serialization, this representation yields more compact sequences. Second, we introduce view-augmented generation, a test-time augmentation procedure that applies axis-aligned rotations to the input mesh, maps all predictions back to a common frame, and selects the final skeleton based on mesh coverage and consistency among predictions from different views. Experiments show that our method achieves better skeleton prediction accuracy than state-of-the-art methods. In particular, our method reduces the CD-J2B error by 16.9% on the Articulation-XL2.0 dataset compared to the strongest directly comparable baseline, Auto-Connect. Qualitative results on in-the-wild meshes further demonstrate generalization across diverse inputs.

The Fourier-cosine (COS) method of Fang and Oosterlee (2008) recovers densities and computes expectations semi-analytically from characteristic functions (ch.f.s). In higher dimensions, both expansion synthesis and COS coefficient-tensor construction scale exponentially with dimension. We address this curse of dimensionality with two tensor-train (TT) methods. A straightforward TT decomposition of the COS coefficient tensor gives COS-TT, which removes the exponential cost of online synthesis but not of offline decomposition. Our main contribution, COS-TT-CHF, instead compresses a ch.f. sample tensor and maps it to the original Fourier–cosine coefficient tensor through an alternative COS representation. Apart from the model-dependent cost of one ch.f. evaluation, both offline and online costs scale linearly with dimension, while error grows only algebraically. COS-TT-CHF introduces two additional errors: frequency-domain truncation and discretization errors. Both are controlled by parameter-selection rules from theoretical error analysis. We further propose an approach for expectations of nonseparable functions of additive scalar aggregates by computing the ch.f.s of the aggregates using tensorized COS methods, reducing the original multidimensional problem to a one-dimensional COS calculation. Closed-form formulas are derived for the TT-contraction integrals arising when the aggregate is a weighted sum, as in European basket option pricing. Experiments under geometric Brownian motion (GBM) and variance gamma (VG) show that, with the parameter-selection rules, COS-TT-CHF remains accurate through 150 dimensions for GBM and 100 for VG within a 30-minute offline computation budget on a laptop. As a by-product, our Fourier-cosine TT constructions yield cosine-basis functional TTs (FTTs) under weaker assumptions than in spectral FTT literature.

Code review is a key quality checkpoint between AI-generated code and production. As AI coding agents submit pull requests at scale, it is unclear whether reviewers reduce scrutiny with repeated exposure and whether review comments reveal this change. We study 11,429 reviews from 400 repeat reviewers over 207 days, paired with 10,104 human-authored inline comments from AIDev. Approval rates rise from 30.5% in reviewers' early periods to 36.6% in late periods (Wilcoxon p = 8.6 x 10^-8; Cohen's d = 0.25). However, four hand-crafted linguistic features - lexical diversity, Shannon entropy, technical specificity, and constructive actionability - show no monotonic decline across exposure deciles (all Spearman absolute rho <= 0.53, p >= 0.11; Bonferroni-corrected Mann-Whitney p >= 0.36). A logistic-regression classifier based on these features reaches F1 = 0.485, below the majority-class baseline. Sentence-embedding structure does carry signal: reviewers' late-period comment centroids shift farther from their early-period centroids than under within-reviewer random permutations (Wilcoxon p < 0.001), and a small MLP using three embedding statistics reaches F1 = 0.74 under 5-fold reviewer-stratified cross-validation. Granger analysis shows that approval-rate changes predict later shifts in technical specificity at all tested lags (p < 0.001), while the reverse direction is significant at only one of four lags. Reviewer adaptation is therefore detectable in latent distributional structure rather than in classical lexical metrics, and language shifts follow rather than precede changes in approval behavior.

LLMs have achieved remarkable capabilities in generating language teaching slides. However, a critical mismatch persists between visual polish and actual instructional effectiveness. To address this gap, we introduce SLATE (Slide-based Learning Assessment for Teaching Effectiveness), the first benchmark that evaluates AI-generated language teaching slides through instructional effectiveness and learner knowledge acquisition. SLATE transforms linguistics olympiad puzzles from low-resource languages with negligible web presence into 90 standardized instructional units comprising 1,133 assessable items, paired with a structured course outline and matched near- and far-transfer test sets. This pretest-posttest design eliminates pretrained knowledge leakage, ensuring gains reflect learning rather than prior recall. Using VLMs as scalable learner proxies and directionally supported by a three-system human pilot, our results show that content validity exhibits a weak association with learning gain, while pedagogical design exhibits a robust positive association. Moreover, most systems show a significant gap between near- and far-transfer accuracy, and even frontier models can produce negative learning gains. SLATE reveals a dissociation between artifact quality and instructional effectiveness, calling for a paradigm shift in how generative teaching systems are built, evaluated, and deployed.

The very first construction by J. Richard Büchi himself for complementing a Büchi automaton relies on a fundamental lemma about the division of an arbitrary infinite word into consecutive finite words, which was cleanly proven by invoking a specialized theorem of Ramsey. For that reason, constructions of similar nature have subsequently been labeled as Ramsey-based. Nevertheless, it suffices to have a weaker form of the lemma where the finite words come from a finite number of congruence classes, rather than arbitrary classes, that form a partition of the set of all finite words. The weaker lemma, with support of nicer properties from a congruence, can be proven without Ramsey's theorem. A commonly adopted improvement on such complementation constructions also requires the working of a congruence. This paper recounts the history and reviews using more contemporary terminology wherever possible the relevant concepts and results, to advocate renaming of Ramsey-based constructions as congruence-based constructions.

Signed distance fields (SDFs) are a popular implicit representation of geometry. Converting a discrete set of SDF samples into an explicit surface is a fundamental problem in geometry processing. Traditional reconstruction methods such as marching cubes and dual contouring ignore the geometric information carried by samples far from the surface. Recently, Sellán et al.[2023] and several follow-up works leveraged the tangent-sphere structure of SDFs; every sample implies a point on a sphere tangent to the surface. However, these approaches extract the zero-level set via surface reconstruction, which considers only points and normals on the surface and ignores the remaining samples. We propose an approach that marries the tangent-sphere observation with radial basis function interpolation of all data, the implied surface points and the original data. By detecting spheres with extremely constrained tangent points, a configuration geometrically forced at sharp surface features, we identify and preserve surface corners that surface reconstruction-based methods systematically round. A partition-of-unity decomposition allows our method to scale efficiently to large grid resolutions. Our reconstructions improve both Chamfer and Hausdorff accuracy at every tested resolution.

Understanding human attention is fundamental for scene interpretation, yet existing approaches often rely on heavily trained models that lack interpretability. Prior methods struggle to jointly reason about gaze targets, attended objects, and visual grounding without extensive supervision. To the best of our knowledge, this work introduces the first training-free Gaze Target Agent (GTA) for gaze-guided reasoning across tasks such as gaze target prediction, attention localization, and object identification. This is achieved by leveraging pretrained vision-language models, augmenting them with visually guided prompts, and employing a memory-based retrieval strategy for high-uncertainty samples to improve performance without additional training. We evaluate our approach using both quantitative metrics and qualitative results. Quantitatively, our method achieves state of the art performance on the GazeFollow and GazeHOI benchmarks. Qualitatively, our agent provides detailed semantic predictions, predicts the correct targets even when ground truth labels are wrong, and remains flexible without vocabulary constraints.

Image-to-video generators are often credited with absorbing physical dynamics as implicit world models, a claim the community currently checks with plausibility scores that ask whether a clip looks consistent with real-world motion. Plausibility is the wrong test on its own, because a clip can look natural while encoding the wrong value of the governing physical parameter, and no existing benchmark measures this gap directly. PhysWeep closes it with a fixed, label-free audit, treating a frozen generator as a black box, recovering the realized parameter from generated pixels, and reporting how often generation is trackable at all, how far the realized value sits from the requested one, and which, if either, of the literature's two proposed failure mechanisms the data support. A deterministic-simulator positive control confirms every score is exactly checkable. Applied to three open generators across six sweep axes, PhysWeep finds a specific, reproducible, previously undocumented failure. Conditional on producing trackable motion, two of the three generate confident, well-fit dynamics that converge to one of a small number of fixed, wrong values selected by the sampling seed rather than by the request, reproducing across two independent model families, two physical systems, and an independent tracker. It matches neither the prior reversion nor the case-based clamping the literature anticipates, because the reversion target is seed-conditional rather than a single global default, and a leave-one-out selection rule rejects both; the in-range faithfulness slope is statistically indistinguishable from zero wherever a response is estimable at all. A benchmark averaging over seeds would never see this: each sample is confidently locked to a wrong constant, exactly the failure a plausibility score is structurally blind to. We release the protocol, suite, and analysis code.

Data-driven techniques have shown promising potential for checking behavior of complex systems operating in safety-critical domains against safety and other temporal requirements. This paper studies a class of data-driven techniques that are based on learning a representation of the system from data using non-parametric estimation. The proposed approach is able to formally verify discrete-time stochastic dynamical systems against temporal logic specifications only using observation samples and without the knowledge of the model, and provides a probabilistic guarantee on the satisfaction of the specification. We first consider finite abstract representations of the system in the form of Markov decision processes (MDPs) and derive asymptotic convergence guarantees between the transition probabilities of the abstract MDP and their estimation using Bernstein's inequality and statistical properties of non-parametric estimators. We then propose theoretical results for estimating the asymptotic upper bound of the Lipschitz constant (LC) of the stochastic system, which can determine the size of the finite abstract MDP for a given precision error. Under appropriate assumptions, our results prove that the asymptotic convergence rate of the estimations is $O(n^{-1/(3+d)})$ for both the transition probabilities and the LC, where $\mathsf d$ is the dimension of the system and $n$ is the data scale. By integrating these results, we can guarantee the asymptotic closeness in formal verification and policy synthesis performed on the original system and its finite abstraction based on the size of the dataset. Multiple case studies are presented to validate the effectiveness of the proposed method.

Channel charting is a self-supervised framework that learns low-dimensional spatial representations from high-dimensional channel state information. We revisit channel charting from a graph-theoretic perspective, and show that the position-diffusion objective is equivalent to a graph Laplacian smoothness functional. We propose a Graph Neural Network (GNN) formulation that replaces the Siamese network's global geodesic dissimilarity objective with a graph smoothness objective. We consider the reformulated objective to be the position diffusion objective. Without diverging from the original optimization objective, the GNN replaces the quadratic-cost self-correlation encoding with linear-cost message passing over an Angle-Delay Profile (ADP)-similarity graph, achieving comparable positioning accuracy with 512 times fewer parameters. Using Laplacian spectral analysis, we demonstrate that obstacles compress the ADP graph spectrum, while the GNN acts as a spectral decompressor against obstacles and other environmental semantics, but as a compressor against excessive ADP embedding space. Beyond this, the second eigenvector of the learned embedding encodes the line-of-sight/non-line-of-sight boundary rather than spatial coordinates.

We study policy optimization for discrete-time robust $\mathcal{H}_\infty$ control with static output-feedback, and present the first feasibility-preserving algorithm with a deterministic, non-asymptotic complexity guarantee. This problem naturally leads to a nonsmooth and nonconvex optimization over the set of stabilizing feedback gains. We first establish several structural properties of the $\mathcal{H}_\infty$ cost. In particular, we show that the cost is weakly convex on every convex subset of a sublevel set. For the state-feedback case, we further establish a weak Polyak–Łojasiewicz inequality, which ensures that every stationary point is globally optimal. Building on these properties, we develop a proximal bundle method for $\mathcal{H}_\infty$ policy optimization. The proposed method can be viewed as an implementable approximation of the proximal point method and uses only function value and subgradient information. We show that all iterates remain stabilizing and establish a deterministic non-asymptotic complexity bound of $\mathcal{O}(\max\{\eta^{-4},\epsilon^{-2}\})$ for finding an $(\eta,\epsilon)$-stationary point. Numerical experiments illustrate our theoretical results.