Papers

The problem of modular scattering-based design of dissipative networks is addressed. Under basic assumptions imposed on subsystems, design of a dynamic scattering-based interconnection controller is proposed which guarantees that the resulting interconnection is internally stable and possesses the same set of basic properties as the individual subsystems. This enables iterative application of the developed technique, which allows for building of large-scale dissipative networks in a modular fashion. The proposed method provides substantial flexibility in the controller design as well as in the choice of input and output signals used for interconnections.

DC microgrids are low- or medium-voltage networks designed to connect and manage DC-based sources and loads. A key challenge in operating DC microgrids is maintaining the DC voltage within certain predefined limits while ensuring its stability. Droop control, the most common method towards addressing this challenge, enables decentralized voltage control and power sharing, but suffers from poor transient performance, resulting in voltage dips and overshoots in applications with fast varying loads such as AI datacenters. This paper introduces an improved droop control method based on voltage-locked loop synchronization, which ensures DC voltage stabilization with significantly improved transient response and achieves the desired load sharing between the available source converters. The core design principle is reflected in the functional separation of the control scheme into a virtual DC machine (VDCM) that operates as a spinning wheel and a virtual current source connected in parallel. The transient response is provided by the VDCM to damp and stabilize DC bus voltage variations, while the slow droop response is provided by the virtual current source to ensure steady-state power balance. The performance of the proposed method is validated in an experimental microgrid setup.

Speech enhancement (SE) models degrade under domain shifts and have to adapt to unseen target domains during deployment. Most existing test-time adaptation (TTA) methods for SE do so by adapting a subset of the model weights using a self-supervised loss, which requires backpropagation at test-time and permanently alters the model. We instead recalibrate the prediction itself and propose BLINC, a training-free TTA method that remaps the predicted time-frequency mask onto a bimodal target distribution by histogram matching. At test-time, the target distribution is parameterized from blind features of the noisy recording, so neither a reference distribution from a classical algorithm nor online metric optimization is involved. BLINC improves the overall quality of both evaluated SE models on almost every target condition and matches or exceeds the loss-based TTA baselines at minimal overhead.

Secondary control and the State-of-Charge (SoC) balance control are important control objectives for battery energy storage systems (BESSs). In this brief, a communication weight allocation method based on the centrality of eigenvectors is designed for a connected and directed graph, which results in the adjacency matrix having a given eigenvector. Subsequently, a distributed secondary voltage controller and an SoC balancing controller are designed for droop-controlled BESSs to achieve voltage leader-following multiconsensus and SoC balancing, respectively. It is worth mentioning that under the designed voltage secondary control scheme, only a single leader is needed to achieve voltage multiconsensus control. In addition, the capacity information/droop coefficient does not need to be transmitted in the communication network to achieve power sharing according to capacity and SoC balance. For SoC balance control, the control gain is also well analyzed to ensure stability. The relevant simulations verify the effectiveness of the designed scheme.

Long transcripts are costly inputs for downstream NLP systems and often contain irrelevant context. We study query-conditioned topic localization: predicting the sentence span in a transcript that best addresses a topic-title query. To improve span localization, we reuse ASR encoder states as sentence-level representations and fuse them with textual embeddings. This lets lightweight span locators exploit speech information without running a separate audio encoder. Experiments on two public datasets show consistent gains over text-only baselines, especially under strict boundary-matching criteria. Cross-dataset experiments further indicate that the benefits are strongest for structured or semi-structured speech, while gains on spontaneous speech are limited and mixed.

This paper presents a novel framework integrating Bayesian physiological modeling with optimal control strategies to achieve uncertainty-aware, personalized hemodynamic regulation during fluid resuscitation. An uncertainty-aware variational autoencoder state-space model (UVAE-SSM) was first developed to capture the dynamical relationship between mean arterial pressure (MAP) and fluid infusion using limited data, while explicitly modeling aleatoric uncertainty (i.e., randomness in the measurements, such as sensor noise). Then, a Bayesian nonlinear state-space model (BNSSM) was developed by utilizing Bayesian neural networks (BNNs) to capture epistemic uncertainty arising from physiological and patient-specific variability, enabling the creation of a virtual patient generator (VPG). Building on this uncertainty-aware modeling framework, a stochastic radial basis function model predictive control (sRBF-MPC) algorithm was designed to track the MAP target while satisfying physiological constraints. Finally, an online fine-tuning algorithm was developed to adapt the nominal UVAE-SSM using streaming VPG data, enabling progressive personalization during closed-loop therapy. Simulation results across unseen animal subjects and an independent human clinical dataset demonstrated the strong predictive accuracy and cross-population generalizability of the UVAE-SSM and BNSSM models. Closed-loop evaluations confirmed that the proposed sRBF-MPC framework achieved stable MAP regulation while providing better risk-aware control compared to quadratic MPC (Q-MPC) and stochastic quadratic MPC (sQ-MPC). Overall, the proposed framework accounts for inter- and intra-patient variability through online model adaptation, offering a promising step toward uncertainty-aware, personalized hemodynamic modeling and control in critical care.

In many delivery settings, a receiver may already hold a lower-quality or lower-resolution representation of an image, obtained through an independent transmission. Conventional codecs encode a subsequently requested higher-quality representation without exploiting this incidental side information, whereas conditional codecs generally assume a prescribed source of side information that is always available. We instead consider an opportunistic setting in which side information may or may not be present. We introduce a single entropy model that conditions on a previously decoded latent when available and falls back to a standard hyperprior otherwise. The proposed adapter maps the side-information latent to the prior signal required by the entropy model, allowing the same model to support multiple target and side-information quality combinations. The analysis and synthesis transforms remain frozen, enabling retrofitting of an existing learned codec while preserving its latent representation and reconstruction path. When the receiver holds the quality immediately below the target, the proposed method reduces the rate of the subsequent transmission by up to 46%, or by 52% when an additional hyper-latent is transmitted. In the absence of side information, the rate penalty remains below 4%, and the reconstructions are bit-identical across the conditional and fallback modes.

In classification-oriented adaptive sensing, posterior samples characterize uncertainty at the current measurement state and can serve two roles: they may guide the next sensing direction, while their class labels provide votes for the candidate classes and determine whether sensing should continue. We focus on the stopping layer that turns these votes into a declaration, without modifying the posterior sampler or sensing directions. A natural plug-in rule declares when the observed vote share exceeds a threshold. We show that this threshold is not itself a confidence guarantee: when the underlying vote mass equals the threshold, the plug-in rule declares about half the time. As alternatives, we calibrate a fixed-sample rule and a finite-horizon sequential rule to a prescribed false-declaration probability, and study exact curtailment, which stops a fixed-pool rule once its final verdict is forced. We then derive how one-round declaration probabilities determine posterior-sample cost and classification accuracy along a sensing path. On MNIST with DDRM and a fixed PCA-guided probe sequence, curtailment saves up to 62% of posterior samples. Among the evaluated rules at matched operating points, sequential stopping reduces the cost the most. At a high accuracy, that same sequential rule can trade more posterior samples for fewer measurements.

Recent advances in diffusion models have enabled high-performance, instance-adaptive compressed sensing through posterior sampling, without task-specific policy training. Existing methods select sensing probes by maximizing total posterior signal variance and are therefore primarily reconstruction-driven. We introduce a classification-driven extension motivated by the closed-form posterior covariance of a class-conditional Gaussian mixture model, which decomposes into within-class and between-class uncertainty. Using calibrated soft classifier outputs, we estimate these uncertainty terms from diffusion posterior samples and propose a classification-oriented criterion for selecting the dominant sensing direction in the unmeasured subspace. Experiments on MNIST and CIFAR-10 compare the resulting classification accuracy, measurement cost, and reconstruction quality with those of reconstruction-oriented counterparts. The results identify regimes in which semantic posterior uncertainty yields a more favorable classification–measurement trade-off and quantify the associated reconstruction cost.

Modeling stiff string vibrations is challenging due to their dispersive and high-frequency characteristics. This study investigates the effectiveness of Physics-Informed Neural Networks (PINNs) in simulating the transverse vibration of a one-dimensional linear stiff string with sharp initial conditions induced by plucking. The governing Partial Differential Equation (PDE), along with the associated initial and boundary conditions, is incorporated directly into the loss function of the neural network. For the reference measurements, a wire-breaking experiment was performed to excite the string, and its vibration response was captured using a laser profiler. The comparison between experimental measurements, finite-difference time-domain (FDTD) simulations, and PINN-based simulations shows good overall agreement, highlighting the potential of PINNs for modeling stiff-string vibrations.

In this paper, we prove that a mean system utilization smaller than one is a necessary condition for the feasibility of real-time systems. Such systems are defined as stable. Stable systems have two distinct states: a transient state, followed by a steady-state where the same distribution of response times is repeated infinitely for each task. We prove that the Liu and Layland theorem holds for stable probabilistic real-time systems with implicit deadlines, we provide an analytical approximation of response times for each of those two states and a bound of the instant when a real-time system becomes steady.

Reconfigurable intelligent surfaces (RISs) are poised to revolutionize wireless multicasting by enabling extended coverage and reliable operation in obstructed environments. These benefits, however, can be undermined by security vulnerabilities arising from practical deployment factors. This work addresses three such critical factors, (i) the discrete nature of RIS phase shifts, (ii) the presence of colluding eavesdroppers, and (iii) the inefficiency of static illumination beams, each threatening security if not properly accounted for in system design. To mitigate these issues, we formulate a joint resource allocation problem that minimizes the wiretap signal-to-noise ratio (SNR) across all eavesdroppers by co-optimizing the RIS configuration and the base station (BS) beam management. This yields a complex, nonconvex mixed-integer nonlinear program (MINLP), which we equivalently reformulate into a tractable mixed-integer quadratically constrained program (MIQCP) solvable to global optimality. Numerical results confirm that the proposed scheme significantly bolsters security, suppressing the wiretap SNR by up to 58% compared to existing baselines.

Fully autonomous tractor–trailer systems are increasingly deployed in logistics, agriculture, and industrial environments, where precise and robust path-tracking capabilities are essential. However, the articulation between the tractor and the trailer introduces additional nonlinearities and significantly complicates lateral and longitudinal control, particularly during reversing maneuvers. This paper introduces a novel path-tracking algorithm specifically designed for articulated vehicles with a single trailer. The proposed method combines a lateral control law applied at the trailer level with a short-horizon predictive adjustment of the tractor steering angle, ensuring stable convergence toward the desired path in both forward and backward motion. The approach is geometry-based and requires no per-vehicle calibration or training. Simulation studies in a high-fidelity physics simulator demonstrate the ability of the controller to match or outperform classical and state-of-the-art methods in terms of accuracy, stability, and robustness to disturbances.

A robot can predict failure and still be unable to prevent it. By the time a safety mechanism reacts, the nominal plan may already have spent the control authority that recovery requires, and fixed task priorities may block whatever response remains. Our key insight is that both aspects are decided inside the controller. Recoverability must inform actions while they are chosen rather than veto them afterward, and task objectives must be adapted as recoverability shrinks. Building on this, we present RAYA, a hybrid learned-analytic framework that places a learned finite-horizon recoverability margin inside an optimal controller with hard constraints and pairs it with a bounded learned scheduler that shifts task weights to facilitate recovery. Across 7,200 simulation episodes per controller spanning quadrotor and autonomous-vehicle benchmarks, RAYA not only improves survival rates, but also transfers the learned components zero-shot to unseen trajectories, disturbances, plant shifts, and friction layouts. We developed an embedded realization of RAYA and deployed it on-board a 35g Crazyflie quadrotor. Across 40 combined hardware flights under wind with either aerodynamic mismatch or an unmodeled 40% motor-command loss, each of three baselines fails in all trials, while RAYA completes 10/10 six-cycle missions. Project Website: https://raya-control.github.io/.

We present the Spoken Wikipedia Presentation Corpus, an extension of the Spoken Wikipedia Corpora featuring LLM-generated slide decks for multimodal ASR. Slides are created from LLM-segmented sections using a hybrid pipeline that combines LLM-based content planning with rule-based design decisions. For each section, an LLM generates a slide title, bullet points, a takeaway message, and a visual description that is used to create an illustration. Rule-based matching then selects layouts, themes, and styles to produce the final slides. A vision LLM extracts slide text as Markdown. We evaluate multiple ASR and spoken language models (SLMs). The best model achieves an average micro-WER of 10.23% and an average micro-CER of 6.48% on audio-only inputs. English yields the lowest error rates, followed by German and Dutch, while performance declines across lower-resource languages. Although audio-only baselines are strong, multimodal zero-shot prompting of omni models remains challenging. The aligned slide, text, and audio data show a strong potential to improve recognition through cross-modal context.

The success of Large Audio Language Models has driven the development of massive multimodal networks exceeding billions of parameters. However, the demand for privacy-preserving, low-latency processing has shifted focus toward Small Audio Language Models (SALMs) capable of on-device execution. In this paper, we introduce Samsone, a family of SALMs designed for edge computing. Our core model, Samsone-134M, establishes a new state-of-the-art for its size class across multiple benchmarks. We further explore the scaling laws of SALMs by introducing Samsone-99M and Samsone-356M. Despite their compact footprint, the Samsone family delivers performance competitive with models orders of magnitude larger. To foster open research and reproducibility, we train Samsone on publicly available data. We release the training code, model weights, mobile-optimized checkpoints and provide an open-source Android application to demonstrate real-time on-device inference of Samsone.

Uniform spherical arrays provide full-space coverage. However, owing to directional element patterns, full-array transmission activates many elements that contribute little to a given user, resulting in increased system overhead and inefficient power allocation. To address this issue, this letter proposes an adaptive element activation strategy for multiuser uniform spherical arrays that reduces the number of physically active elements while satisfying individual user rate requirements. The proposed strategy constructs a user-specific candidate sequence according to the angle between each element boresight and the user direction. At each iteration, it selects, among the unsatisfied users, the user-element connection that provides the largest improvement in the aggregate rate deficit. Simulation results show that fewer than half of the elements are sufficient for a single user to achieve 80% of its full-array rate, while fewer than 70% are sufficient to achieve the full-array rates of all users in multiuser scenarios. Moreover, the proposed strategy consistently requires a lower active-element ratio than the baseline methods across all tested configurations.

Tabla Stroke Transcription (TST) converts tabla audio into symbolic stroke sequences, but the scarcity of annotated recordings makes fully supervised training challenging. We propose a semi-supervised framework that uses sequence-level labelled and unlabelled tabla recordings. A teacher model generates pseudo-label sequences, while a Stroke-Level Confidence Estimation Model (S-CEM) estimates confidence for each predicted stroke. To improve training with uncertain pseudo-labels, we propose Confidence-Guided Markov Weighted Alternative Temporal Classification (CMW-ATC), which weights candidate sequences within uncertain spans using learned stroke transitions and reliable neighbouring strokes. Experiments across three TST evaluation settings show consistent improvements over conventional teacher–student training and Alternative Temporal Classification in its replacement form (ATC-R). Ablation studies further show the contributions of S-CEM, Markov transition weighting, and the use of reliable strokes on both sides of uncertain spans.

Cooperative multi-monostatic sensing enables sub-meter passive localization in integrated sensing and communication~(ISAC) networks by fusing base station~(BS) observations at a central unit~(CU). Existing studies, however, treat the fused data as ideally available at the CU, overlooking how the per-BS 3GPP functional split jointly constrains fronthaul bitrate, computational load, and whether coherent or non-coherent fusion is feasible at the CU. We propose the Split-Aware Joint Resource Allocation for Multi-Monostatic Sensing~(SARAMS) algorithm, which minimizes the worst-case multi-target squared position error bound~(SPEB) by jointly optimizing per-BS power, bandwidth, and observation time under fronthaul, computational, and power constraints. A split-dependent coefficient embeds the feasible fusion type into the Fisher information matrix~(FIM), casting SPEB minimization as a mixed-integer nonlinear program (MINLP) solved via a semidefinite program for power allocation and block coordinate descent (BCD) over a dominance-pruned configuration set. Simulation results demonstrate that SARAMS reduces the 90th-percentile worst-case SPEB by $65.7\%$ over equal power allocation while attaining an $8.7\%$ optimality gap relative to exhaustive search.

Feedforward Spiking Neural Networks (SNNs) typically propagate every generated spike indiscriminately, disregarding whether the information is redundant from an information-theoretic perspective. This lack of selectivity induces high redundancy in inter-layer communication, creating an expensive overhead, e.g., in scenarios involving many-core neuromorphic hardware or communication-dominated Internet-of-Things (IoT) where features are transmitted wirelessly. To address this challenge, we trade localized processing for leaner network channels by introducing a minimal predictive coding framework for SNNs. We propose two layer variants sharing a predictor block: error units, which transmit signed spiking residuals, and predictive suppression, which uses residual magnitude to dynamically gate and forward only unpredictable, "surprising" activity. Evaluated on the N-MNIST and Spiking Heidelberg Digits (SHD) datasets using diagnostic metrics that decouple local processing from cross-layer communication, our new predictive coding layers achieve significant communication savings. Numerical results reveal a three-fold reduction in communicated activity, while increasing the task accuracy for both datasets. The latter finding is notable, and suggests that predictive coding layers not only minimize communication overhead, but also produce output feature vectors with a higher representation power.

Audio deepfake forensics is moving beyond a simple real-or-fake verdict toward attribution: which system generated the audio clip? Most source-attribution methods cast this as closed-set classification, so they cannot name a generator that was absent from training, a gap that widens with every newly released text-to-speech (TTS) system. We instead frame attribution as cross-modal retrieval: each generator is described in natural language, and a clip is attributed by retrieving the description closest to it in a shared audio-text embedding space. Adding a new system then takes nothing more than writing its description, with no retraining and no new classifier head. Our model couples a frozen Wav2Vec2-BERT audio encoder with a frozen E5 text encoder and aligns them through small trainable projection heads, using a contrastive objective that combines a cross-modal supervised-contrastive loss with intra-modal terms. We evaluate on MLAAD v9 (140 TTS models, 51 languages) under 10-fold leave-models-out cross-validation. For generators it has never encountered before, the model reaches a model-level mean reciprocal rank (MRR) of 58.4%. Even when the correct model is not identified, the audio clip is often matched to systems that share the true generator's vocoder, acoustic model, or architecture. Because the same embedding space also answers natural-language attribute queries, one set of descriptions covers both open-set attribution and attribute-level forensic profiling.

Fully-actuated multirotor aerial vehicles must not only track nominal wrenches but retain the "readiness" to modulate them rapidly under disturbances. Classical effort-minimizing allocators ignore this dynamic limit, whereas maximizing readiness leads to topologically disconnected optimal sheets demanding physically impossible actuator rates. Enforcing a readiness safety floor on fixed-geometry symmetric platforms further encounters a zero-sum degeneracy: motor-speed redistribution cannot improve authority without conceding wrench tracking. This paper uses active morphology to break the degeneracy, treating servo tilt as a geometric resource supplying authority-recovery directions unavailable to static rotors. We construct a configuration-dependent, motor-only readiness certificate - the log-volume of the reachable wrench-rate set - that explicitly excludes servo capacity, preventing a "ghost capacity fallacy" in which the certificate would falsely credit slow mechanical kinematic limits instead of collapsing accurately at motor saturation. The certificate is enforced as a Control Barrier Function (CBF) within a Unified Physical-Command Quadratic Program acting on motor torques and servo setpoints. Closed-loop simulations of an articulated octorotor under severe gust disturbances show classical allocators diverging and uncertified articulated allocators violating the safety floor, while the proposed CBF filter bounds the system state and preserves vehicle authority.

Although memristor-based in-memory computing (IMC) has been widely investigated for brain-inspired neuromorphic workloads, systematic evaluations of its energy, latency, and signal-to-noise ratio (SNR) trade-offs across diverse system parameters remain scarce for classical digital signal processing (DSP). To address this gap, this paper introduces a comprehensive hardware-aware design framework for systematically mapping algorithmic workloads onto IMC architectures, using the discrete Fourier transform (DFT) as a case study. Tailored to the stringent performance requirements of a DFT accelerator for 5G orthogonal frequency-division multiplexing (OFDM) systems, we propose a novel mapping scheme that reduces energy consumption per computation by 53% compared with conventional mapping techniques. Additionally, we present a comparative study of resistive random-access memory (RRAM) and ferroelectric tunnel junction (FTJ) technologies with identical numbers of programmable states but distinct conductance ranges, demonstrating the importance of device-informed architectural co-design. By bridging the gap between emerging IMC architectures and rigid DSP constraints, this work provides a pathway toward energy-efficient edge accelerators for future wireless communication systems.

WiFi-based human activity recognition has achieved high accuracy under the single-user scene. Recognizing activities performed by multiple concurrent users remains challenging, because their body-reflected propagation paths superimpose in the channel state information (CSI) measurement. Existing multi-user methods either decompose the signal before classification, suffering from error propagation, or predict activities jointly without reliably associating each prediction with the correct user. Moreover, attention mechanisms that have proven effective for single-user sensing compute a single user-agnostic attention map, blending patterns from different users and providing no mechanism to associate features with specific user slots. This paper proposes a user-conditioned spatial attention (UCSA) module that generates per-user spatial attention maps by conditioning multi-scale spatial attention on learnable user-ID embeddings via feature-wise linear modulation (FiLM). UCSA creates a deep coupling between spatial feature enhancement and user differentiation: the attention module itself becomes user-aware, producing distinct attention maps for each user slot rather than relying on a late-fusion embedding addition. Combined with a shareable multi-semantic spatial attention (SMSA) module for multi-scale spatial enhancement, the proposed framework addresses both feature quality and user differentiation in a unified architecture. Experiments on the WiMANS benchmark across three indoor environments and up to five concurrent users show that the proposed method outperforms nine baselines in all settings, achieving an average accuracy of over 93% across three environments even with five users. Ablation studies confirm that both SMSA and UCSA contribute substantially to recognition accuracy, with UCSA providing the critical link between shared features and per-user predictions.

Residential heating accounts for a large share of building energy use, and predictive control strategies can reduce it by anticipating rather than reacting to the room temperature alone. The usual manner of implementing predictive control, model predictive control (MPC), requires an accurate model of each individual residential unit for heating control. Obtaining this model is time-consuming to obtain manually or even impossible, and such models do not transfer across a heterogeneous building stock. Data-enabled predictive control (DeePC) removes this modeling step by designing the controller directly using measured trajectories of the system. The foundations of DeePC, however, are built on the class of deterministic linear time-invariant systems, which is an unrealistic assumption for residential heating. This work applies three recently proposed nonlinear extensions of DeePC, collectively referred to as data-driven predictive control (DPC): Select-DPC, gain-scheduling DPC, and linear parameter-varying DPC. These are compared against standard DeePC and the hysteresis controller, the industry standard in residential heating. The comparison runs over a full heating season on a calibrated digital twin of an occupied research unit, the NEST research building in Switzerland, assessed by heating energy and comfort-band violation. Moreover, gain-scheduling DPC was deployed on the real residential apartment and validated the conclusions from the simulations. Each of the DPC controllers consumes significantly less energy than the hysteresis controller, amounting to roughly $11\%$ over the season, validating the use of DPC. The nonlinear methods further outperform linear DeePC in all assessed scenarios, and we analyze the relative costs and benefits of each nonlinear method.

We define OmniVChat (Omni Video Chat) as the task of native audio-visual dialogue between a user and an omni model. In OmniVChat, omni models directly and simultaneously receive audio and video from a user and return text. The user's query is embedded in the audio and video, without a separate text question, external captioning, or speech recognition. Direct audio-visual input reduces external latency and computation while preserving perceptual cues. However, research on OmniVChat faces two constraints: data availability and evaluation. Recordings of people using their own devices are scarce. Furthermore, good replies often depend on multimodal context and can be phrased in many ways, making keyword matching unreliable for evaluation. Recent progress in agent systems and video generation makes generation for comprehension viable, which means using synthesized dialogues for training and evaluation. Therefore, we present OmniVChat-Studio, a multi-agent data engine for synthesizing single- and multi-turn audio-visual dialogues. We use synthesized dialogues to build OmniVChat-Bench, an evaluation benchmark that evaluates omni models' basic dialogue abilities across five ability categories. Replies are judged by a large language model based on explicit scoring criteria. We also present OmniVChat-RL, a reinforcement learning reward design that jointly targets reply correctness, efficiency, and style in OmniVChat. Training Qwen3-Omni-Instruct with OmniVChat-RL on synthesized dialogues improves its performance on both OmniVChat-Bench and the human-recorded OmniVChat-Bench-Human. These gains validate the reward design and show transfer to real-world dialogues in training and evaluation.

Humanoid locomotion over complex terrain requires anticipating footholds that may no longer be visible at touchdown. Limited camera coverage and self-occlusion make it necessary to retrieve relevant terrain information from earlier observations. We present FootQuery, a perceptive locomotion framework that queries depth history using each foot's predicted next touchdown. The policy predicts touchdown locations and uncertainty from proprioception and uses these distributions, together with per-foot features, to query sparsely sampled historical depth frames. During training, realized contacts are projected into historical images to supervise retrieval at the regions where those contacts were visible. The retrieved per-foot features are fused with global visual memory to generate control actions. A progressive force-assistance curriculum supports early exploration, while event-consistent tread-midline shaping encourages coordinated stair contacts. Deployment requires only proprioception and onboard depth images. In simulation, the complete framework outperforms its component ablations on the most challenging tested stairs, gaps, and platforms. Real-world experiments on a Unitree G1 demonstrate continuous traversal with a single policy across outdoor stairs and indoor routes combining stair ascent and descent, platforms, and gaps. These results support organizing visual history around anticipated contacts for perceptive humanoid locomotion.

The growing integration of DERs underscores the importance of unified coordination across TSO, DSO, and aggregator levels. This paper proposes a coordination method between TSO, DSOs and DER aggregators, based on a transparent, well-scalable, and decentralized single-loop iterative mechanism with limited information sharing. The approach enables iterative price-based coordination between adjacent levels, where the higher-level entities provide price signals and the lower-level ones respond with desired power schedules. The proposed method uses the proximal-gradient-based iterative co-optimization algorithm extended to a three-level case, which enjoys convergence characteristic of proximal algorithms. The performance of the proposed approach is compared with both a centralized formulation and a sequential reference formulation based on fixed transmission-level energy prices. The results demonstrate that the proposed coordination achieves outcomes very close to those obtained with the centralized formulation, while significantly improving performance relative to the sequential reference, particularly at higher levels of DER flexibility.

Neural Radiance Fields (NeRF) have attracted extensive attention in recent years due to their strong capability for high-quality 3D reconstruction and novel view synthesis from multi-view images. Existing methods usually rely on high-quality sharp inputs, while real-world image acquisition is highly susceptible to blur degradation, which severely affects the reconstruction quality of NeRF. In this paper, we propose a novel Mamba-driven world-state-aware adaptive deblurring neural radiance field, termed WS-NeRF, to address image degradation and 3D inconsistency. We formulate the alternating optimization of radiance fields as a dynamic evolution process with temporal memory, and jointly exploit comprehensive multi-dimensional world states and a mixture-of-experts mechanism to dynamically adjust the confidence of deblurring priors. Experimental results show that WS-NeRF significantly improves blurry radiance field reconstruction quality, achieving better performance on PSNR, SSIM, and LPIPS, while exhibiting more stable iterative recovery behavior.

In this paper, we revisit the convergence of the RLS algorithm in view of the Cauchy Interlace Theorem. It states that the eigenvalues of a Hermitian matrix $A$ and of the rank one updated matrix $\hat{A} = A + u\,u^H$ interlace. We analyze the impact of this theorem on the convergence of the RLS algorithm and its suitable initialization. The results suggest an improved initialization for speeding up the convergence time. Moreover, a novel proof of the theorem is given based on the Sherman Morrison Woodbury formula.