Papers

Accurate path loss (PL) estimation is essential for wireless network planning and optimization. This paper proposes HankelKANet, a lightweight physics-inspired network for PL estimation that structurally embeds electromagnetic propagation characteristics into the neural architecture. Built on a single residual UNet backbone, the proposed model integrates urban layout features with a convolutional Kolmogorov-Arnold network(KAN). Specifically, the convolutional kernels are parameterized by the radial amplitudes of the zeroth- and first-order Hankel functions. This formulation acts as a physics-inspired nonlinear distance encoding, enabling the network to capture dominant attenuation characteristics directly without requiring additional physics-based loss terms. Experimental results demonstrate that HankelKANet achieves state-of-the-art accuracy and efficiency on both the simulated RadioMapSeer dataset and the real-world Huawei RSRPSet. Compared with the leading baseline, PEFNet, it reduces the parameter count by 78% and lowers the RMSE by 16.96% on the simulated data, while further exhibiting robust cross-dataset transferability and high data efficiency during target-domain adaptation. To interpret these empirical gains, we mathematically establish a parameter-conditional spatial gradient bound. Using discrete Dirichlet energy (DDE) and Jacobian sensitivity heatmaps, we empirically verify that this physical encoding intrinsically constrains the network's spatial derivatives. This structural smoothness effectively eliminates non-physical spatial ripples, preventing overfitting to simulator artifacts and serving as the fundamental driver for robust real world generalization.

Robotic exploration, coverage, and multi-agent deployment algorithms require reproducible environments with controllable geometry and visibility. This paper presents the Dungeon and Visibility World Generator, an open-source framework for constructing two- and three-dimensional environments and evaluating geometric line-of-sight. The framework supports orthogonal, polygonal or polyhedral, and curvilinear geometries, with six construction families covering scattered rooms, clusters, winding passages, caverns, branches, and loops. Room dimensions, passage structures, obstacles, and spatial placement can be controlled through configurable parameters and probability distributions. Each generated environment provides continuous free-space geometry and a corresponding discrete occupancy representation. A Rust-based visibility engine supports line-of-sight queries in both representations, enabling the evaluation of observable regions from arbitrary agent positions. The framework records configurations, generated geometries, occupancy grids, and visibility results to support reproducible experiments. We demonstrate the generator across all six geometry classes, examine how geometry and spatial resolution affect visibility, and evaluate generation time and reproducibility. The resulting framework provides a configurable foundation for evaluating robotic exploration, coverage, and multi-agent deployment algorithms in diverse environments.

Cameras and the foundation models built on them recognize what a 3D scene contains, but not what it is made of: surfaces of the same colour can be different materials. Hyperspectral unmixing explains each spectrum as a mixture of material signatures, yet in 3D it is applied to rendered images after the views are composited, or tied to slow implicit fields, and no multi-view data with known composition exist to check the result. We present the Spectral World Renderer (SWR), which unifies world generation and unmixing. A coding agent builds executable 3D worlds that are rendered with measured spectra, so every hyperspectral image comes with exact labels and sub-pixel proportions; the resulting SWR-Bench provides eight indoor and outdoor worlds in which decompositions can be verified. SWR then unmixes inside rendering: each Gaussian of a compact 3D representation is decoded into material abundances before compositing, so one render yields spectra and material maps, and any surface can be queried by material name, a clicked pixel or a reference spectrum. Without material labels, SWR identifies materials best in all eight worlds, has the best mean PSNR on real captures, trains 3.7-12.9x faster than unmixing baselines, renders 34.9 labelled views/s and separates same-coloured materials.

In multi-modal 3D pre-training, aligning 3D shapes, their 2D image counterparts, and language descriptions has made tremendous progress. However, we find that supervision from rendered images is unreliable, the semantics encoded in synthetic views are distorted due to domain gap, self-occlusion, and rendering artifacts. Instead, we learn transferable 3D representations from language only. We observe that, in real world, 3D shapes are organized in a tree-like structure, ranging from general concepts to fine-grained instances with specific attributes, while language typically denotes generic concepts and subsumes these various 3D instantiations. Hence, we propose to embed 3D representations in hyperbolic space, whose geometry is well-suited for modeling tree-like data without distortion. To capture the asymmetric hierarchy between language and 3D shapes, we further impose a partial-order relation that language entails 3D shapes. Our method, Hyperbolic Contrastive Learning (HyperCL), jointly optimizes a hyperbolic geodesic-based contrastive objective and an entailment loss to learn discriminative cross-modal representations while explicitly preserving this hierarchy. We also introduce a rigorously verified benchmark for comprehensively evaluating cross-modal 3D alignment. Across two widely used benchmarks and ours, we improve the performance up to 4.6% absolute percentage on 3D shape classification and 5.5% on cross-modal text-to-3D retrieval under the same parameter budget. The code and evaluation benchmark will be made publicly available.

While multi-fingered dexterous robot hands have the potential to provide human-level dexterity, using them for functional manipulation remains difficult because task success depends on precise contact, coordinated arm-hand motion, and reactive recovery from execution errors. Precise contact depends on complex contact dynamics that are difficult to model accurately in simulation, motivating real-world learning. However, real-world demonstrations in the full arm-hand control space are costly and noisy because each trajectory must coordinate arm motion with high-DoF multi-finger contact. Small execution deviations can further cause slipped contact or premature force application, making recovery from off-nominal contact states important for robust execution. To deal with these challenges, we propose HiFun, a hierarchical real-world RL framework for functional dexterous manipulation tasks. Our key insight is to decouple the control complexity into a functional hand skill for precise contact execution and an arm-hand coordination policy for arm control and hand-skill activation. This design enables effective real-time human intervention for reactive recovery learning and efficient data collection with Dynamic Movement Primitive (DMP) augmentation. Across six real-world tasks on two dexterous hands, HiFun achieves 98.3% average success within one hour of online training and remains robust to dynamic disturbances. The project page is available at https://hly-123.github.io/HiFun/.

Vision-language-action (VLA) models connect visual perception and language understanding with robot control, providing a foundation for instruction-conditioned manipulation. However, conventional pretrained VLA policies lack a mechanism for gradual adaptation through accumulated interaction experience during deployment. We therefore propose VLRLA, a general framework that equips existing VLA models with in-context reinforcement learning capabilities for deployment-time adaptation. Our approach augments a VLA model with a reinforcement learning module that learns from historical observations, actions, rewards, and termination signals. During deployment, the policy gradually refines its action generation and progressively improves task success while keeping model parameters fixed. Our approach is also designed for general applicability and compatibility with diverse learning history construction strategies and model training pipelines. Finally, through extensive experiments, we systematically investigate the learning properties of VLRLA in simulation and demonstrate its ability to adapt to unseen scenarios during real-world deployment.

In this paper, we investigate a wireless communication network supported by multiple uncrewed aerial vehicles (UAVs). To ensure reliable connectivity, UAVs are desired to approach the boundaries of designated no-fly zones (NFZs) without violating them when mobile users (MUs) enter NFZs. To support heterogeneous communication services, MUs are assigned different priority levels based on their individual communication types. Considering the practical constraints arising from user mobility, heterogeneous communication requirements, and NFZs, the objective is to maximize the weighted sum of MU communication rates, where weights correspond to MU priorities. This is achieved by jointly optimizing trajectory planning, UAV transmission power control, and the binary association factor between UAVs and MUs. To address this challenging optimization problem under dynamic constraints, we develop a soft actor-critic algorithm with prioritized experience replay, which learns an adaptive control policy. Numerical results show that the proposed algorithm outperforms benchmark schemes in terms of convergence speed and stability. Moreover, it effectively meets diverse communication requirements, satisfies UAV energy constraints, and ensures reliable operation under NFZ constraints.

We present a CR-CR method for strain gradient elasticity and prove its robustness with respect to the Lame parameter $\lambda$ and the length scale $\iota$. The method introduces the strain as an independent symmetric tensor and discretizes both the strain and the displacement using CR elements. The commuting properties of the CR interpolation operator with the relevant differential operators leads the robustness with respect to $\lambda$. A stabilization term weakly imposes the relation between the strain and the symmetric gradient of the displacement. The method also applies directly to the biharmonic equation and singularly perturbed fourth-order problems. For these problems, we prove that the CR-CR method is equivalent to an RT-CR method proposed by Huang and Tang. This result extends the classical equivalence between the RT-P0 mixed method and the CR method for the Poisson equation to fourth-order problems. The equivalence shows that the stabilization term arises naturally from the RT-CR mixed formulation.

Feel-good Thompson sampling (FGTS) augments the posterior of vanilla Thompson sampling with an exploration term that favors reward functions with high predicted optimal values. This modification achieves minimax-optimal regret up to logarithmic factors in linear contextual bandits, but existing analyses of its two standard variants establish regret guarantees in expectation. In this paper, we derive the first high-probability regret bounds for both cumulative and current-step FGTS in contextual bandits with general realizable function approximation. Our bounds depend on the corresponding decoupling coefficients and the prior mass near the true reward function. In the $d$-dimensional linear setting, both variants satisfy $\mathrm{Regret}(T)=O\big(d\sqrt T\log(2T)+\sqrt T\log(1/\delta)\big)$ with probability at least $1-\delta$, where $T$ is the time horizon. This bound matches the dimension and horizon dependence of algorithms based on the upper confidence bounds, up to logarithmic factors. Relative to the expected-regret guarantees, the cost of controlling the regret with high probability is confined to additive terms with at most linear dependence on $\log(1/\delta)$. The proof ideas are partially developed by GPT-6 Astra under the authors' guidance.

Block-diffusion language models denoise a whole block of token positions in parallel so each forward pass can emit many tokens. However, blocks are still denoised in strict sequence, a cost that within-block acceleration leaves untouched. Here, we investigate decoding blocks concurrently and propose One Block Ahead (OBA), a training-free decoding algorithm that denoises the next block alongside the current one. We demonstrate that for uniform-state diffusion LMs, where a block's denoising progress is not visible in its intermediate tokens, concurrent decoding is advantageous if the next block's tokens are promoted together with the corresponding sampler state, including the self-conditioning signal and the denoising time. We also derive a break-even criterion that relates the benefit of OBA to the relative cost of a paired-block forward pass. On DiffusionGemma, a state-of-the-art diffusion LM, OBA increases the tokens per forward pass by 42.3% (from 15.64 to 22.25) and the single-request denoising throughput by 16.3%, averaged over ten benchmarks, with a small accuracy regression.

We propose CktHGVAE, a heterogeneous graph variational autoencoder for transistor-level analog/mixedsignal CMOS circuits that takes not only design variables but also state variables as its inputs and outputs. With a single hyperparameter configuration without topology-specific tuning, comparison between model-generated and SPICE-simulated circuit operational states gives median R2 values above 0.9 for typical single-ended voltage amplifiers and the CMOS inverter, and 0.81-0.96 for differential amplifiers with larger datasets. Latent-space optimization in the PCA-reduced latent space with Bayesian optimization, evolutionary search, random search, and quasi-exhaustive search is demonstrated for a common-source voltage amplifier with NMOS input and resistive load, where the objective function is defined as a penalty for deviation from required circuit specifications. For each search strategy, 93.3 - 100% of all design candidates suggested by the model satisfy the required circuit specifications at the decoder output, and SPICE simulations verify that 90 - 100% of them indeed satisfy the requirements. The model pretrained on a batch of topologies can be transferred to unseen topologies without retraining at all, with zero-shot reconstruction median R2 values of 0.72-0.90 in three of four scenarios, including transfer from single-ended to differential topologies, although the transfer fidelity depends on the choice of pretraining topologies. Our demonstrations suggest that the proposed model can potentially serve as a basis for AMS CMOS circuit design automation.

An agentic scientific workflow can run to completion and still release wrong results: a silent numerical error, a stale input, or a defect in a shared upstream step can contaminate several outputs while a sibling branch stays correct. We treat trust in such results as a release decision. A harness around the workflow's directed acyclic graph (DAG) decides which outputs may be used: it enforces mandatory checks, contains failures to the affected branch, recomputes from tracked versions, adds validators, and releases a result only if a risk model estimates its probability of being invalid as low. Language-model agents work inside the harness, choosing corrective actions and reviewing releases, but cannot override it. On 60 held-out paired instances of two workflows with injected faults, we switched components on and off. Each supplied a different property. Containment and recovery raised the number of correct results. Validation supplied safety: fixed rules without validation released 37 invalid results, while validating every candidate released none. Risk estimation supplied efficiency, matching that safety with 31% fewer validators at somewhat lower yield, although better targeting than random validation was not shown. Lineage evidence helped on one workflow. The language model added no measurable benefit: whether invalid results escaped depended on the harness, not the model. This proof of concept points to a practical path: placing agents inside explicit release-control structure can keep released scientific results trustworthy, and it gives more capable agents a measurable baseline to improve on.

In the classical art gallery problem, a guard sees a target when the segment joining them lies entirely within the polygon. We explore what changes when sight may bend along regular arcs on absolutely irreducible curves, but only those defined by primitive integer polynomials of total degree at most \(d\) and coefficient height at most \(H\). We call such curves admissible sight carriers. Arcs can dramatically improve sight. In a one-sided comb, no straight-line guard sees selected targets in two different teeth, yet a single central guard sees a chosen target in every tooth along parabolic carriers. Sight can also fail in two different ways. Geometrically, the polygon may block every sight arc of the allowed degree between two points. For instance, in an alternating comb, the number of guards needed for a row of targets spanning the comb grows linearly with the number of sides at each fixed degree, even for carriers with arbitrary real coefficients. Arithmetically, no admissible carrier may pass through the points at all. Even in the unit square, at fixed \(d\) and \(H\), the proportion of rational targets that must guard themselves tends to one as the height bound grows. An algebraic target may lie on no rational curve of degree at most \(d\), however large \(H\) is. Both failures can be overcome for finite target sets by allowing the degree and coefficient height to grow. We show that every finite set of algebraic targets in a rational polygon becomes visible from any rational interior guard, with explicit bounds. We also construct, for every \(d\), arbitrarily large Galois orbits in the unit square, all of projective height \(4^{d+1}\), whose points must each guard themselves at degree \(d\) for every \(H\). Yet at degree \(d+1\) and coefficient height one, a single guard sees the whole orbit along \(Z(Y-X^{d+1})\).

This paper proposes the SKG (Symplectic-Kernel Gradient) reasoning engine for law-induced inference and induction. Laws are a priori; data provide instances. Inference is geometric evolution under law constraints. We organize symplectic geometry, GENERIC, kernel methods, RKHS, group-equivariant networks, and constrained mechanics into a unified, analyzable, implementable, verifiable framework. We propose the SKG octuple S = (M, g, omega, H, S, G, Delta, Pi), whose four components correspond to four levels of laws: conservation laws, constitutive relations, symmetry groups, and axiom systems. Main results: (1) Invariants enter metric construction via the Jacobian: g = J^T Sigma^-1 J + epsilon_G I. For constant cases, we establish inverse reconstruction theory: solvability rank condition (Thm 3.9), consistency criterion (Thm 3.11), and uniqueness theorem (Thm 3.14). (2) The kernel bandwidth matrix is uniquely determined by Fisher information and satisfies task-independence (Thm 3.19). (3) The metric reconstruction error decomposes into four terms (Thm 3.21), with squared condition number kappa(Sigma^*)^2 as the dominant amplification factor (Thm 3.22). (4) For state-dependent cases, we establish function-space reconstruction theory on matrix-valued RKHS (Thms 4.8, 4.9). (5) Three-layer projections handle axiom, symplectic-step, and conservation constraints. (6) The residual equation yields the observable formula, four-dimensional rank condition, and noise propagation, while g_qq is structurally unidentifiable. (7) Kernel parameterization yields the RKHS extrapolation bound (Thm 4.15) and strict OOD gain theorem in group directions (Thm 8.15). (8) The learning-decoupling theorem (Thm 9.1) proves the sole object of induction is the parameters of the task energy. Validation: 1D damped oscillator and 2D rotation-invariant system; 6 unit tests, 10 experiments (paired t-test, p < 0.01).

Spatial intelligence requires scalable 3D environments that are both photorealistic and semantically grounded, yet existing indoor datasets are dominated by room-scale captures and costly LiDAR acquisition. Online real-estate platforms offer abundant panoramas of diverse multi-room homes, but sparse viewpoints and texture-poor regions make reconstruction severely under-constrained. We introduce Holi-Spatial 360, a generative reconstruction and data curation framework for converting these panoramas into annotated multi-room 3D environments. Because reconstruction quality bounds subsequent generation, our Correspondence-Guided Geometric Optimization converts feed-forward reconstruction priors into reliable cross-view constraints and builds a coherent 3D Gaussian Splatting representation as 3D spatial memory. Adapted single-image and video priors then restore appearance and complete continuous trajectories while preserving observed content. The reconstructed scenes support arbitrary RGB-D rendering and an automated curation pipeline that combines end-to-end 3D grounding with VLM-based verification. Holi360 contains over 1K multi-room 3DGS scenes, more than 100K instance annotations, and 2M spatial question-answer pairs. We also introduce the Holi360 benchmark to evaluate the spatial understanding of VLMs in real-world, multi-room indoor environments. Experiments demonstrate superior reconstruction and curation quality under sparse panoramic inputs, while fine-tuning VLMs on Holi360 substantially improves spatial reasoning capabilities.

Text-to-speech (TTS) based on masked diffusion models (MDMs) achieves competitive zero-shot speech synthesis performance. Unlike autoregressive (AR) models, MDMs define a corruption process that maps clean speech-token sequences to masked maps and learn a reverse denoising process to recover the reference tokens. However, MDMs commit to intermediate predictions that cannot be revised in subsequent inference steps. Consequently, prediction errors accumulate and propagate, degrading synthesized speech quality. We present Revise-TTS, a single-stage framework that applies hierarchical binary quantization (HBQ) and generative refinement networks (GRN) to zero-shot speech generation. First, we replace finite scalar quantization (FSQ) with HBQ to introduce a scalable binary representation for speech tokens with an explicit coarse-to-fine structure. Second, GRN is trained to reconstruct the token sequence from a mixture of ground-truth and random tokens. During inference, iterative refinement enables self-correction by randomly discarding a small proportion of previously predicted tokens with new predictions. Experiments and ablation studies demonstrate that Revise-TTS achieves competitive zero-shot speech synthesis performance using HBQ. It can further be extended to any MDM-based TTS task as an alternative training and inference method.

Model signing authenticates the full-precision weights W an LLM is released with, yet users run a quantized artifact Q whose producing transformation is not uniquely determined under current disclosure. We separate the non-environment inputs from the execution environment E and measure a quantity that, to our knowledge, has not been measured before: the Independent Verifiability Rate (IVR), the fraction of a popularity-sampled GGUF ecosystem that a third party can re-derive bit-for-bit. Across the 200 most-downloaded Hugging Face GGUF repositories (198 in scope), no artifact or model card records the build environment, so under a recorded-inputs protocol the IVR is 0/198. Yet the artifacts are not irreproducible. On four in-sample repositories that disclose their inputs, the card's information alone reproduces no Q4_K_M file, but two coordinates the card omits - the build's floating-point-contraction setting and, for hybrid models, the per-tensor quantization recipe - re-derive 7/8 target-quant files bit-for-bit on x86 (5/8 on arm64); the fp16 re-derives from the provider's checkpoint, but the calibration imatrix must be shipped and hash-bound, since recomputing it did not match. An older release additionally pins the instruction set and did not re-derive. Reproducing the divergence reported with llama.cpp PR #25353, we show that E is a necessary coordinate and that one compiler setting yields a unique output across the tested platforms at negligible quality cost (+0.06% perplexity). As a security characterization, benign floating-point non-determinism separates from uniformly placed nibble-flip tampering above a detection floor that grows with the environments a verifier must allow, but a targeted embedding edit evades the statistic, so the proposed protocol rests on exact re-derivation; adversarial-fp16 attacks lie outside it. We release the audit data and a re-runnable harness.

Data augmentation (DA) plays a vital role in improving model performance in low-resource or limited-supervision scenarios within Natural Language Processing (NLP). Existing DA methods, such as synonym replacement and back-translation, have demonstrated effectiveness on the word or sentence level; however, they frequently neglect discourse-level coherence and logical flow, which are essential for complex tasks dependent on inter-sentential relationships. In this paper, we propose a structure-preserving document-level data augmentation framework based on large language models (LLMs) and document-level discourse structure parsing. Our approach identifies rhetorical relations between sentence pairs and extracts key phrases, which are then replaced with semantically coherent but topically shifted content while preserving the original discourse structure. Experimental results on text summarization and question answering show that the proposed augmented data provides complementary supervision and improves combined-data training across the evaluated tasks. Comparisons with a standard RST parser and controlled ablations further support the proposed discourse representation.

Retrieving useful context for a robot requires matching its task, execution stage, and the continuation that should follow. We introduce Robota-Embed, to our knowledge the first vision-language-action (VLA) embedding model to jointly encode all three modalities in both current-state queries and recorded future continuations. Predictive retrieval training aligns these endpoints, enabling historical continuations to serve as context at inference time. We construct a multi-source corpus of 1,695,684 cleaned training anchors and introduce Robota-Embed-Bench, with 4,000 queries and a shared 8,000-entry gallery. Robota-Embed achieves 47.33% clip-level and 81.77% episode-level R@1, exceeding action-free Gemini Embedding 2 by 22.09 and 33.47 percentage points. Controlled retrieval studies show that removing native action degrades retrieval more than removing vision. We study two complementary applications. Embed2ICL uses retrieved experience for training-free and training-based policy conditioning, increasing success from 47.0% to 56.3% in DROID-Sim and from 86.4% to 92.8% on the LIBERO difficult subset, while improving task progress on real robots. Embed2Evolve uses execution feedback to grow task-local memory and refine context use around a frozen GPT-6 Astra policy. Across five ManiSkill tasks, it achieves 68% success, compared with 24% for GPT-6 Astra Direct and 33% for a harness independently evolved with Omni-Embed-Nemotron-3B under the same protocol. Relative to the latter, it uses approximately 79.5% fewer policy tokens per test rollout, counting both successes and failures. These results show that a unified, action-aware retrieval space can support robot in-context learning and task-local context evolution.

Real-time monocular human pose estimation is an important sensing component for human-centered intelligent systems. While recent camera-based methods have achieved impressive accuracy, fundamental challenges persist, particularly when visual observations are incomplete, such as during lower-body occlusion or short out-of-frame intervals. In these challenging but common scenarios, vision-only systems often struggle to reconstruct physically plausible lower-limb motion and reliable foot-ground contact. Due to the highly unobtrusive and wearable nature of smart insoles, this paper investigates whether simple plantar pressure and acceleration sensors can effectively resolve these visually induced limitations. To systematically investigate this, we first collect a synchronized multimodal dataset and then propose a part-decoupled temporal fusion framework. The proposed model fuses foot-sensing features into the lower-body branch to compensate for incomplete visual evidence, while a lightweight coordinator exchanges information between the upper- and lower-body temporal representations to maintain whole-body consistency. The framework is evaluated with three representative real-time visual backbones based on 2D keypoints, 3D landmarks, and SMPL-based human mesh recovery. We evaluate our framework on three representative real-time visual backbones. Quantitative experiments demonstrate that incorporating simple foot sensors consistently improves pose accuracy, motion smoothness, and gait-related measurements over vision-only baselines, with the most significant gains observed under occlusion and short visual-dropout conditions. Ultimately, our results demonstrate that unobtrusive foot sensing provides vital and complementary physical observability when camera observations are temporarily unavailable.

Streaming Inverse Text Normalization (ITN) is vital for converting spoken-form outputs from streaming Automatic Speech Recognition into formatted written text. Existing streaming ITN methods rely on hybrid systems combining neural tagging with expert-crafted finite-state transducer rules, limiting scalability across domains and languages. While end-to-end models offer superior scalability by learning directly from data, standard encoder-decoder architectures are inherently non-streaming due to global attention. In this paper, we propose an efficient streaming end-to-end ITN system, adapting a pretrained text-to-text model to leverage its robust linguistic knowledge. To enable streaming, we introduce architectural adaptations, a specialized training strategy, a Read-Tag-Write decoding policy, and inference optimizations. Experiments on Vietnamese datasets show accuracy comparable to non-streaming baselines, outperforming hybrid methods while satisfying real-time latency requirements.

The rise of quantum computing presents substantial challenges to the security of Peer-to-Peer (P2P) energy trading systems. This paper introduces an innovative mutual authenticated session key agreement protocol aimed at protecting edge-IoT P2P energy trading communications in the quantum era. Each peer is an IoT-enabled smart metering device that is registered by an edge registration authority and acts as both producer and consumer. The proposed protocol incorporates CRYSTALS-Kyber's quantum-resistant cryptographic suite to guarantee the authenticity and confidentiality of negotiated session keys in the presence of quantum threats. Furthermore, it leverages the Hedera Hashgraph public blockchain, which is recognized for its quantum-resilient design, to ensure transparency and support efficient revocation mechanisms. Extensive comparative analyses have been conducted to demonstrate the protocol's superiority over existing alternatives, including both traditional schemes based on classical cryptography and emerging early-stage post-quantum approaches, in terms of performance and security features. By combining robust security and high efficiency, the proposed protocol fosters the development of resilient decentralized energy markets, paving the way for secure and sustainable energy distribution in the future.

Digital-domain self-interference cancellation is essential in in-band full-duplex radios after passive suppression and analog cancellation. It is particularly challenging when the desired signal is strong relative to the residual self-interference. In this regime, conventional iterative digital cancellers update their coefficients from desired-signal-contaminated received samples and may suffer substantial performance degradation. This paper reformulates digital-domain self-interference cancellation as an operator-learning problem. The proposed framework learns the temporal evolution of the residual self-interference channel from historical observations and uses the learned operator together with the known local transmit signal to reconstruct the current residual self-interference. A Transformer neural operator is used to realize this framework. The resulting method is trained in a desired-signal-free setting and then applied in the presence of a desired signal. A continuity-based analysis characterizes the sensitivity of the learned operator to the resulting input perturbation. Numerical experiments in static and time-varying channels, including time-varying SIR conditions, demonstrate stable cancellation performance over a wide range of interference levels. At 30 dB SIR, the proposed method improves the cancellation performance by 5–15 dB over the considered conventional methods.

Sparse matrices, in which most entries are zero, are central to scientific simulation, graph analytics and machine learning, and their multiplication with dense inputs is a core parallel kernel. In many applications the matrix stays fixed while only a few input rows change per step, so the result can be updated from those rows alone, which we call the dirty set. Sparse kernels often add a second optimization based on the inspector-executor principle, in which a tiled schedule is prepared once and reused across calls. However, when this reuse remains profitable as the dirty set moves is not well characterized. In this paper, we show that retained tiled execution rarely beats direct CSC in incremental sparse matrix multiplication, although incremental execution itself remains highly profitable. Across 21 SuiteSparse matrices and controlled patterns of motion, a direct column-wise incremental update is 2.1-8.2x faster than the best of eight full-recomputation codes, among them Intel oneMKL and PyTorch. Yet the same direct update outperforms a tiled executor that retains tile preparation and caches assembled schedules in 779 of 837 workloads. Separating reuse into per-tile preparation and cached schedules shows that the benefit comes from preparation; extra cache capacity seldom helps because a tile set rarely returns after a different one has intervened. Tiling is faster only when tiles merge many entries into each output write, a structural property that we predicted before measurement and confirmed on four held-out matrices of the same family. These results show that the structure of the matrix and the motion of the dirty set, rather than the sophistication of the schedule, determine whether reuse is profitable. We anticipate that our methodology and guidelines will help developers decide when to build schedule reuse into incremental sparse runtimes.

We examine the value of information in extensive-form decision problems: the largest multiplicative gain that a decision maker can obtain from an information refinement, taken over all payoff environments. This measures the structural importance of information disentangled from any particular payoff specification, and generalizes the notion of value of recall recently introduced by Berker et al. (AAAI'25). We establish a unifying characterization of the value of information under any information refinement, uncovering an intriguing connection to the maximum fractional chromatic number over a certain family of conflict graphs. We then employ our characterization to elucidate the complexity of two central information refinements. First, for the value of perfect information in perfect-recall decision problems, we find that the relevant conflict graphs are cographs, so the fractional chromatic number collapses to the clique number. Even though the clique number is tractable for a given cograph, we show that the additional layer of optimization (over the family of graphs) makes computing the value of perfect information NP-complete. On the positive side, we prove fixed-parameter tractability in terms of the degree of imperfect information. Second, for the value of recall in imperfect-recall decision problems (with no absentmindedness), we first show that the value of recall can realize the fractional chromatic number of any graph. Interestingly, this implies that, unlike the value of perfect information, the value of recall can take fractional values. Building on this reduction, we then establish S2P-completeness. This addresses questions left open by Berker et al. (AAAI'25)—both on the characterization and the complexity of the value of recall—and shows that computing the value of recall is strictly harder than the value of perfect information.

Mandlebrot's coastline paradox reframes measurement as a joint property of an object and its ruler. Here we ask how observational resolution shapes the apparent structure of neural computation. We introduce a scale-indexed decomposition that separates the predictive information captured by direct dependencies from the residual outside that model family. Exact constructions show that projection can convert higher-order dependence into pairwise structure. Projection can increase the direct fraction of surviving information even as total predictive information decreases. A schematic branch-to-soma cascade grounds these effects in local coincidence detection, saturating integration and somatic mixing. Resolving organized nonlinear subunits improves approximation and raises effective representational rank. In a nested hierarchy, these gains occur without increasing information about the target. This distinction separates information availability from access by a restricted readout. A point-neuron or direct population model therefore establishes an effective description at a specified scale. It does not uniquely identify the finer computation that generated the observed relation. Which resolution is adequate depends on the observable and perturbation of interest.

Velocity space discretization is a primary source of computational cost in deterministic methods for multiscale gas flows. A fixed velocity grid often retains redundant kinetic degrees of freedom (DOFs), particularly under near equilibrium conditions. To address this issue, this work develops a nonequilibrium-capturing formulation based on the unified gas kinetic scheme (UGKS), in which the number of kinetic DOFs adapts locally to the degree of nonequilibrium. A discrete orthogonal basis is constructed for numerical implementation, and the velocity distribution function is decomposed into an analytical local equilibrium and a basis expanded nonequilibrium component. The retained, locally adapted DOFs are evolved directly, substantially lowering computational overhead while preserving the discrete conservation of mass, momentum, and energy, as well as accurately capturing nonequilibrium states. The formulation retains the core multiscale coupling of molecular transport and collision inherent to UGKS, providing explicit and implicit schemes for unsteady and steady flows, respectively. Numerical tests across a wide range of flow speeds and rarefaction regimes (from continuum to highly rarefied) demonstrate close agreement with benchmark solutions. For unsteady flows, the explicit scheme achieves speedups of 12.64 to 58.50 over the standard UGKS. In steady state tests, only about $2\%$ of the original kinetic DOFs are required in continuum regions, accelerating each implicit iteration by a factor of 33.87. Even in the most rarefied cases, the scheme captures nonequilibrium behavior using only about $27\%$ of the original DOFs.

Sequential conformal prediction needs to decide which past errors should calibrate uncertainty for the current target. Recency and online error feedback are useful when recent observations remain relevant, but the recent past can be a poor reference when comparable states recur at nonadjacent times, with the recurrence pattern generally unknown. We propose State-Matched Sequence Conformal Prediction (SMSCP), which constructs an explicit one-sided local-linear state from pre-target information, uses state similarity to select a matched reference set of past calibration scores, and then applies the ordinary unweighted conformal order statistic. In contrast to existing approaches, SMSCP uses state similarity to identify calibration observations relevant to the target score quantile. This motivates a theoretical analysis of whether nearby states have comparable target score quantiles under temporal dependence. We establish control of the resulting conformal cutoff and characterize predictive information not captured by the chosen state. Simulations show that state matching is most effective when score-relevant states recur across nonadjacent times, while revealing settings in which its advantage weakens or reverses. Across solar irradiance and bike-demand benchmarks, SMSCP improves interval score over recency, weighted, adaptive, and HopCPT comparators, and achieves coverage closer to the nominal 0.90 level than ResCP on five of six benchmarks.

Wrist-worn inertial measurement units (IMUs) provide a common sensing modality for tasks ranging from activity and gesture recognition to pose estimation, inertial odometry and clinical motor assessment. Existing wearable benchmarks evaluate performance across multiple tasks, but leave open how supervision for one task affects transfer to others and whether these tasks benefit from a shared representation. We present IMU-Bench, a benchmark of ten tasks from eight datasets for studying IMU multi-task performance and cross-task transfer. We compare six self-supervised methods pretrained on unlabeled recorded and simulated wrist motion, evaluating classification, pose and velocity estimation, and IMU-language retrieval. We find that our masked autoencoder, InrtlWrist, achieves the best or joint-best frozen-probe performance on nine of ten tasks, and fine-tuning improves on training from scratch on nine of ten tasks, including full-body pose, odometry and retrieval. Across transfer and joint-training experiments, supervision that benefits one target can reduce performance on others, with supervised sources transferring within kinematic estimation while, under frozen probing, none approaches the pretrained encoder on language retrieval, and joint-training gains that come at a cost to the paired task. These findings distinguish broad downstream usefulness from complementary task supervision. Code, task cards and a leaderboard are publicly available.

Sim-and-real co-training is an effective way to scale robot learning by augmenting limited real-world demonstrations with simulation data. However, existing approaches typically treat simulators as fixed data sources, relying on uniform generation, offline retrieval, or manual parameter alignment by domain experts. We formulate Active Generative Simulation: optimizing a simulator's data-generation distribution to maximize downstream real-world performance. To solve this problem, we propose MetaGenSim, a meta-learning framework that trains an auxiliary rater network via meta-gradients to automatically discover high-utility data-generation distributions within a user-specified simulation parameter space, without manual parameter alignment. Across simulation and real-world manipulation tasks, MetaGenSim outperforms uniform generation and retrieval baselines, achieves a 150% relative improvement in real-world full-task success over uniform generation, and matches expert manual alignment. We further show that distributions curated with a lightweight proxy policy transfer to VLA fine-tuning, and that MetaGenSim can exploit partial alignment through atomic skill stitching and object-receptacle composition when exact sim-and-real task alignment is unavailable. Project website: https://meta-gensim.github.io