arXiv subjects

Yan Yang

Publications and source records attributed to Yan Yang.

At least 19 recordsLinked to original sources

arXiv2026

DADKC: Distribution-Aware Adaptive Differentiated Knowledge Consolidation for Cloth-Hybrid Lifelong Person Re-identification

Lifelong Person Re-identification (LReID) aims to match the same person using sequentially collected data. In practical long-term deployments, however, the model must handle both domain shifts across cameras and scenes, and irregular clothing changes of the same identity. This gives rise to the Cloth-Hybrid Lifelong Person Re-identification (CH-LReID) task, which demands balancing cloth-relevant and cloth-irrelevant knowledge in streaming data. While existing LReID methods have explored replay-based or distillation-based anti-forgetting strategies, they typically fail to sufficiently model stage-aware domain distributions and cross-stage feature continuity under exemplar-free training. To address this problem, we propose a Distribution-Aware Adaptive Differentiated Knowledge Consolidation (DADKC) framework. The core idea is to enhance the existing Differentiated Knowledge Consolidation (DKC) paradigm with adaptive distribution modeling. To this end, three lightweight modules are designed: (1) an Anchor Pool Adapter (APA) to retrieve learnable anchor cues for refining identity features; (2) a Phase Path Mixer (PPM) to perform stage-aware expert routing for integrating domain-specific distribution paths; and (3) a Stage Bridge Alignment (SBA) module to regularize the continuity of base, anchor, and path representations between adjacent stages. Extensive experiments on the CH-LReID and LReID benchmarks show that DADKC gains 4.1%/4.5% mAP/R1 on Order-1 and 3.6%/3.4% mAP/R1 on Order-2 over the DKC baseline, and lowers weighted forgetting from 2.4%/1.7% to 1.8%/1.4%, while storing neither historical images nor per-image features. The code is available at https://github.com/fire667/DADKC.

cs.CV↗

arXiv2026

From Transcription to Reformulation: Defining and Benchmarking Grounded Generative Speech Recognition

Speech language models have expanded automatic speech recognition (ASR) from transcription to general speech-to-text generation. Yet existing benchmarks still assume a single ground-truth transcript per utterance, an assumption that struggles when models produce readable, formatted, or structured outputs. In this work, we introduce Grounded Generative Speech Recognition (GSR), reformulating speech recognition as generating semantically grounded text instead of a single transcript. We define a grounded reformulation space that characterizes all valid textual realizations supported by the input speech, and establish a four-level hierarchy of reformulation capabilities, including literal transcription, canonical reformulation, semantic reformulation and speech-grounded structuring. Based on this formulation, we present GSR-Bench, the first comprehensive benchmark designed to evaluate generative speech recognition systems beyond exact transcript matching. GSR-Bench covers diverse scenarios including refined meeting transcription, cross-dialect translation, context-aware speech understanding, and robustness under silent or noisy acoustic conditions. To address the limitations of word-level evaluation metrics, we develop a multi-dimensional evaluation framework combining speech grounding assessment and LLM-based semantic judgment, whose reliability is validated through human preference studies. Extensive experiments across representative architectures, ranging from conventional ASR systems to end-to-end Speech-LLMs, reveal fundamental trade-offs between transcription fidelity and generative reformulation ability. Our analysis further identifies critical failure modes, including acoustic hallucination, unsupported generation, and instruction drift, providing insights into the design and evaluation of future grounded generative speech systems.

cs.CL↗

arXiv2026

Efficient Maximum Likelihood Estimation of Multiple Membership Linear Mixed Models, with an Application to Educational Value-Added Assessments

The generalized persistence (GP) model, developed in the context of estimating ``value added'' by individual teachers to their students' current and future test scores, is one of the most flexible value-added models in the literature. Although developed in the educational setting, the GP model can potentially be applied to any structure where each sequential response of a lower-level unit may be associated with a different higher-level unit, and the effects of the higher-level units may persist over time. The flexibility of the GP model, however, and its multiple membership random effects structure lead to computational challenges that have limited the model's availability. We develop an EM algorithm to compute maximum likelihood estimates efficiently for the GP model, making use of the sparse structure of the random effects and error covariance matrices. The algorithm is implemented in the package GPvam in R statistical software. We give examples of the computations and illustrate the gains in computational efficiency achieved by our estimation procedure.

stat.AP↗

arXiv2026

FluxVLA Engine: A One-Stop VLA Engineering Platform for Embodied Intelligence

Vision-language-action (VLA) models, world-action models (WAMs), and offline reinforcement learning methods are rapidly expanding the design space of embodied policies, yet turning these algorithms into reliable robot systems remains constrained by fragmented data formats, training stacks, evaluation protocols, inference runtimes, and embodiment-specific interfaces. We present $\mathrm{FluxVLA}$ Engine, an open, configuration-driven platform that turns heterogeneous embodied-policy components into a reproducible data-to-deployment workflow. Rather than introducing another policy model, $\mathrm{FluxVLA}$ standardizes interfaces for datasets, visual-language and world models, action heads, reward- or advantage-weighted learning, distributed training, simulation evaluation, optimized inference, and robot operators. The engine further integrates compositional dual-arm simulation, scalable automatic data generation, and model-decoupled human-in-the-loop rollout, takeover, correction collection, and reward annotation. For responsive physical execution, it combines Real-Time Chunking (RTC) with accelerated inference backends, lightweight remote GPU serving, and configurable trajectory post-processing. Together, these capabilities connect offline learning, simulation validation, online correction, and real-robot execution through shared and auditable contracts. $\mathrm{FluxVLA}$ therefore targets the engineering bottlenecks separating promising embodied-learning algorithms from reproducible evaluation and dependable deployment. Code is available at https://github.com/FluxVLA/FluxVLA

cs.RO↗

arXiv2026

Computation of Maximum Likelihood Estimates for Multiresponse Generalized Linear Mixed Models with Non-nested, Correlated Random Effects

Estimation of generalized linear mixed models (GLMMs) with non-nested random effects structures requires approximation of high-dimensional integrals. Many existing methods are tailored to the low-dimensional integrals produced by nested designs. We explore the modifications that are required in order to adapt an EM algorithm with first-order and fully exponential Laplace approximations to a non-nested, multiple response model. The equations in the estimation routine are expressed as functions of the first four derivatives of the conditional log likelihood of an arbitrary GLMM, providing a template for future applications. We apply the method to a joint Poisson-binary model for ranking sporting teams, and discuss the estimation of a correlated random effects model designed to evaluate the sensitivity of value-added models for teacher evaluation to assumptions about the missing data process. Source code in R is provided in the online supplementary material.

stat.CO↗

arXiv2026

Omni-Woodpecker: Hallucination Correction via Evidence-Prior Rebalancing for Omni LLMs

Recent Omni LLMs (OLLMs) have demonstrated remarkable capabilities in joint audio-visual reasoning, yet cross-modal hallucinations remain a major challenge. Existing mitigation methods mainly operate at the output level, overlooking the entanglement of grounding evidence and learned priors within fused representations. To address this issue, we propose Omni-Woodpecker (OWP), a training-free framework that shifts hallucination mitigation from output-level manipulation to conflict-conditioned representation rebalancing. Specifically, OWP first identifies cross-modal conflicts from complementary modality-specific views. It then employs orthogonal decomposition to separate grounding evidence from competing priors within the hidden states. Finally, OWP adaptively rebalances fused representations by suppressing misleading priors and restoring ignored evidence before decoding. Extensive experiments demonstrate the effectiveness and strong cross-architecture generalization of OWP. Using Qwen2.5-Omni-7B as the backbone, OWP improves performance by 12.9 and 8.6 points on CMM and AVHBench, respectively, while consistently achieving gains of 0.67-2.25 points across four general OLLM reasoning benchmarks. Furthermore, OWP generalizes effectively to VideoLLaMA2-AV, improving performance by 4.6 and 3.9 points on the same benchmarks. Code is publicly available at https://github.com/ZiiJianLiu/Omni-Woodpecker

cs.AI↗

arXiv2026

Joint Beamforming Optimization and Dynamic Tracking in RIS-Enabled Secure ISAC Systems

This paper investigates a reconfigurable intelligent surface (RIS)-enabled secure integrated sensing and communication (ISAC) system, where the direct links between the base station (BS) and users are blocked, and the mobile eavesdropper is treated as the sensing target. The mobility of the eavesdropper induces time-varying wiretap channels, potentially compromising secure transmission when outdated eavesdropper information is used. To address this challenge, the BS dynamically tracks the eavesdropper over consecutive time slots and exploits the predicted state information to adapt its transmission strategy. A sum secrecy rate maximization problem is formulated by jointly optimizing the BS beamforming, artificial noise, and RIS reflection coefficients, subject to a constraint on the tracking error covariance to ensure the required sensing accuracy. To tackle the resulting nonconvex optimization problem with temporal coupling, an iterative algorithm integrating an extended Kalman filter (EKF) with an alternating optimization framework is developed. Specifically, the EKF recursively predicts and updates the eavesdropper's state, while the joint optimization problem is decomposed into four tractable subproblems that are solved iteratively. Simulation results demonstrate that the proposed algorithm achieves improved secrecy performance while effectively tracking the trajectory of the mobile eavesdropper, compared with benchmark schemes.

eess.SP↗

arXiv2026

Movable Antenna-Enabled Multi-Static Cooperative Secure ISAC: Architecture, Mechanisms, and Enabling Technologies

The broadcast nature of integrated sensing and communication (ISAC) signals exposes them to physical-layer security (PLS) threats, including eavesdropping and jamming. Conventional fixed-position antennas provide limited spatial flexibility under highly correlated legitimate, sensing, and adversarial channels, while multi-static cooperative ISAC offers distributed spatial diversity but remains vulnerable to information leakage. To address these limitations, this article proposes a novel movable antenna (MA)-enabled multi-static cooperative secure ISAC framework centered on dual-scale spatial reconfiguration. Specifically, continuous MA repositioning reshapes local propagation channels, while distributed multi-static nodes provide complementary network-level spatial diversity. Their synergy transforms local antenna geometry and distributed network geometry into a unified controllable spatial resource for jointly enhancing communication, sensing, and PLS. This article first presents the system architecture of the proposed framework and its representative application scenarios, followed by an analysis of the physical-layer mechanisms underlying the resulting gains. Subsequently, we elaborate on the key enabling technologies for practical deployment. A case study further demonstrates the communication-sensing-security trade-off and antenna-saving potential of the proposed framework. Finally, open challenges and future research directions are identified toward practical MA-enabled multi-static secure ISAC networks.

eess.SP↗

arXiv2026

Dual Secure Sensing and Communications in MA-Enabled Networked ISAC Systems

Integrated sensing and communication (ISAC) is envisioned as a cornerstone technology for next-generation networks, yet it introduces significant security risks in both communication and sensing domains. To mitigate the coverage limitations inherent in conventional mono-static or bi-static setups, this paper proposes a networked ISAC framework empowered by movable antennas (MAs). By leveraging flexible antenna reconfiguration, MAs provide additional spatial degrees of freedom to significantly boost beamforming gain and interference suppression. Specifically, we investigate a novel dual-domain security paradigm where networked base stations collaboratively protect uplink (UL) and downlink (DL) transmissions from communication eavesdropping while simultaneously safeguarding target sensing data against sensing eavesdroppers (SEs). To this end, we formulate a joint optimization problem to maximize the sum of secrecy rates, subject to sensing requirements, SE eavesdropping probability constraint, and practical MA deployment constraints, by jointly optimizing receive combiners, transmit beamformers, UL power allocation, and MA positions. To address this non-convex problem with highly coupled variables, an efficient alternating optimization algorithm is proposed, integrating closed-form solutions for receive combiners, successive convex approximation for transmit beamforming and UL power, and a tailored sensing-aware multi-candidate particle swarm optimization for the high-dimensional MA positioning. Extensive numerical results validate the effectiveness of the proposed algorithms, demonstrating that the proposed MA-enabled networked cooperative ISAC design significantly outperforms conventional fixed-position antenna-based designs in terms of security performance, while achieving notable power- and antenna-saving gains.

eess.SP↗

arXiv2026

HWAM: Humanoid World Action Model With Joint State--Action Generation

Humanoid robots are a promising platform for general-purpose manipulation. Recent Vision-Language-Action (VLA) policies learn actions directly from multimodal observations, while World Action Models (WAMs) further incorporate future visual prediction to improve action generation. However, in hierarchical humanoid systems, VLA and WAM policies output reference actions that are subsequently realized through whole-body control, robot dynamics, balance, and contact. This hierarchy creates an action--execution gap: the reference produced by the policy can differ from the motion realized by the robot. Without explicitly modeling the realized body state, future visual prediction must jointly explain scene evolution and discrepancies between reference actions and executed motion, making it difficult to associate an action with its physical outcome. We propose HWAM, a Humanoid World Action Model with joint state--action generation, which makes the robot's post-execution proprioceptive state an explicit prediction target. By jointly generating reference actions and their realized body states, HWAM directly incorporates supervision of executed motion into action learning. HWAM is trained through three complementary conditional paths. The Policy path jointly denoises state--action trajectories conditioned only on current observations, matching deployment conditions. Forward Dynamics Modeling (FDM) predicts future visual observations conditioned on actions and post-execution states, while Inverse Dynamics Modeling (IDM) reconstructs the joint trajectory from visual transitions. Together, these paths connect policy references, realized body motion, and visual outcomes. HWAM achieves the highest success rate among evaluated baselines on three real-robot tasks on the LimX OLI humanoid. On Candy Picking, HWAM achieves a 70.6% success rate, compared with 43.3% for Fast-WAM.

cs.RO↗

arXiv2026

VAttnPO: Visual Attention-Guided Policy Optimization

Multimodal reinforcement learning with verifiable rewards commonly assigns one sequence advantage to an entire response. This outcome-level signal distinguishes overall response quality but leaves changes in visual-evidence use across generated tokens implicit. We introduce VAttnPO (Visual Attention-Guided Policy Optimization), which uses answer-conditioned visual attention to refine token-level credit while preserving the verifier-derived trajectory-mean advantage. A frozen old policy constructs a visual reference by teacher forcing the known correct answer and produces a visual distribution for each rollout token. VAttnPO measures their agreement in visual-attention space using Jensen--Shannon divergence, rather than reading guidance from teacher output-token distributions or introducing an attention-matching objective. Trajectory centering, relative-position weighting, and credit balancing turn this agreement into a zero-mean, position-aware correction to the GRPO sequence advantage. Token intervention, visual-index permutation, and trajectory analysis provide diagnostic evidence linking this signal to model behavior, visual correspondence, and generation position. On Qwen3-VL-8B-Instruct, VAttnPO raises six-task image AVE from 70.97 with GRPO to 73.50 and scores higher on every reported structured, video, and MME metric. Gains over GRPO also hold on Qwen2.5-VL-7B-Instruct and InternVL3-8B-Instruct.

cs.CV↗

arXiv2026

Two-Step Data Augmentation for Masked Face Detection and Recognition: Turning Fake Masks to Real

The absence of large-scale masked face datasets challenges masked face detection and recognition. We propose a two-step generative data augmentation framework combining rule-based mask warping with unpaired image-to-image translation via GANs, producing masked face samples that go beyond rule-based overlays. Trained on about 19,100 images in the target domain (3.8% of IAMGAN's scale), or, including out-of-domain transfer pretraining, 59,600 and 11.8%, the proposed approach yields consistent improvements over rule-based warping alone and achieves results complementary to IAMGAN's, showing that both steps contribute. Evaluation is conducted directly on the generated samples and is qualitative; quantitative metrics like FID and KID were not applied as any real reference distribution would unfairly favor the model with closer training data. We introduce a non-mask preservation loss to reduce non-mask distortions and stabilize training, and stochastic noise injection to enhance sample diversity. Note: The paper originated as a coursework project completed under resource constraints. Following scholarship termination, the author took on part-time employment to maintain research continuity, which led to a mid-semester domain pivot from medical imaging to masked face tasks due to company data restrictions. The work was completed alongside concurrent coursework with delayed compute access and without AI assistance. It was submitted at the semester end to meet a publication requirement, accepted without revision requests, and selected among the best papers for an extended submission to Springer Nature Computer Science, which was not pursued due to continued funding absence. Downstream evaluation on recognition or detection performance was not completed by the submission deadline. The note is added in response to subsequent comparisons and criticisms that did not account for these conditions.

cs.CV↗

arXiv2026

Binding Multiple Modalities via Multimodal Wasserstein Barycenter

Multimodal learning beyond two modalities commonly leverages a specific modality (e.g., text) to bind other modalities. However, how to establish a more balanced representation space that approximates shared semantics while respecting the holistic geometry of $n$-modal data remains challenging. In this work, we present BaryBind, which aims to transport the specific modality towards the Wasserstein barycenter (WB) optimized across all modalities and introduces a volumetric alignment objective to establish a unified semantic space around the WB embedding. Specifically, we project specific modalities to the WB, which minimizes the average Wasserstein distances to multimodal distributions and serves as the anchor for subsequent alignment. We then construct a barycenter simplex, whose volume is taken as a similarity metric for global alignment centered at the WB. Experiments show that BaryBind achieves competitive performance in text-video-audio retrieval, classification, videoQA, and cross-modal generation tasks, along with robustness under modality absence and scalability to more than three modalities. Code is released at https://github.com/xl-tang3/BaryBind.

cs.LG↗

arXiv2026

Generalized Duke's theorem for signed Graphs

Duke's interpolation theorem states that the orientable genera of a connected graph form an integer interval, and Stahl established the corresponding result for nonorientable embeddings. In 1991, Širáň showed that this interpolation property fails for signed graph embeddings: the Euler-genus spectrum of a signed graph may contain gaps. He subsequently asked whether all such gaps must occur at the lower end of the spectrum. In this paper, we establish a characterization of the Euler-genus spectrum of a connected signed graph. We prove that, for each parity class, the Euler genera form a step-two interval. Moreover, whenever both parity classes are nonempty, their maximum elements differ by one. As an consequence, if two consecutive integers $k$ and $k+1$ belong to the Euler-genus spectrum, then every integer from $k$ to the maximum Euler-genus also belongs to the spectrum, thereby answering Širáň's question affirmatively. Our proof uses the pre-signed graph representation of signed embeddings together with ordered adjacent-exchange operations and a matching interpretation of face numbers.

math.CO↗

arXiv2026

A Tutorial on Movable Antenna-Enabled ISAC Systems: Fundamentals, Parameter Estimation, and Security Issues

Movable antenna (MA) technology has recently emerged as a promising paradigm for enhancing the flexibility and performance of wireless systems by allowing antennas to dynamically adjust their spatial positions within a confined region. Meanwhile, integrated sensing and communication (ISAC) has attracted significant attention as a key technology for next-generation wireless networks, aiming to integrate communication and sensing within a shared hardware and spectral framework. By leveraging the additional spatial degrees of freedom offered by MA, MA-enabled ISAC systems provide new opportunities to enhance both communication performance and sensing accuracy. However, the integration of MA into ISAC also poses new challenges in parameter estimation and security. In this tutorial, we provide a comprehensive overview of MA-enabled ISAC systems, with a particular focus on their fundamentals, parameter estimation, and security issues. First, we introduce the basic principles of MA and ISAC, and review representative MA-enabled ISAC systems and their potential applications. Then, we establish unified mathematical models for communication and sensing, and systematically review representative channel and sensing parameter estimation methods tailored to MA-enabled ISAC systems. Numerical comparisons are provided to illustrate the characteristics and performance of different estimation methods. Furthermore, we investigate the security issues of MA-enabled ISAC systems, introduce the fundamentals of physical-layer security (PLS), and summarize how the spatial reconfigurability of MAs can be exploited to enhance the security of ISAC systems. Representative case studies are presented to demonstrate the performance gains of MA-enabled secure ISAC systems. Finally, we identify several open issues and future research directions toward practical MA-enabled ISAC systems.

eess.SP↗

arXiv2026

SocialOmni: Benchmarking Audio-Visual Social Interactivity in Omni Models

Evaluating omni-modal large language models (OLMs) in multi-party dialogue requires more than answer correctness on pre-segmented inputs. We introduce SocialOmni, an offline diagnostic benchmark that separates three turn-level decisions: identifying who is speaking, deciding when a designated participant should enter at an annotated query time, and determining how that participant should continue the dialogue. SocialOmni contains 2,000 perception items and a quality-controlled core split of 200 interaction-generation items, including naturally occurring speaker-visibility mismatches. Each item is independently checked by three human annotators. Evaluated systems receive only query-time-bounded multimodal evidence, while manually verified reference continuations are reserved for response judging. A complete three-judge ensemble scores every eligible response, with leave-one-judge-out and family-sensitivity audits. Across 11 OLMs, rankings vary substantially by axis, and coverage-adjusted scores reveal when high conditional response quality depends on selective turn entry. The protocol does not measure persistent streaming state or wall-clock latency.

cs.AI↗

arXiv2026

SaltyMeta: a curated benchmark and protein language model-informed web tool for salty peptide prediction

Excess sodium intake remains a major public health challenge, while salty and saltiness-enhancing peptides offer a potential route to preserve sensory saltiness in reduced-sodium foods. Machine-learning studies of salty peptides, however, are constrained by small datasets, heterogeneous evidence standards, uncertain negative labels, and sequence similarity leakage. Here we present SaltyMeta, a curated benchmark and web-accessible screening framework for salty or saltiness-enhancing short peptides. The benchmark contains 580 peptides, including 280 positive peptides and 300 negative peptides, all standardized to 2-15 residue one-letter amino-acid sequences. Quality control found no non-standard residues, exact duplicates, or positive-negative overlaps. A similarity-grouped split retained 456 peptides for training and 124 for held-out testing. We evaluated 548 interpretable peptide descriptors, frozen ESM2 embeddings at 8M, 35M, and 150M parameter scales, and descriptor-embedding fusion models under grouped cross-validation. The traditional ExtraTrees baseline selected by training-set grouped cross-validation achieved ROC-AUC=0.693 in cross-validation and ROC-AUC=0.704, PR-AUC=0.702, F1=0.626, and MCC=0.304 on the held-out test set. The best initial protein-language-model fusion was traditional descriptors plus ESM2-8M embeddings, with grouped CV ROC-AUC=0.696 and test ROC-AUC=0.700. Advanced optimization using PCA95 dimensionality reduction and ExtraTrees feature-importance filtering yielded a practical ESM2-8M PCA95 top-300 model with test ROC-AUC=0.715 and PR-AUC=0.703, although repeated-CV gains remained modest. SaltyMeta is a transparent prioritization tool, not a sensory validation substitute. We provide benchmark, models, scripts, GitHub, and Streamlit for reproducible screening of food-derived peptides.

q-bio.BM↗

arXiv2026

Rapid Loss of the Sierra Nevada's Largest Trees Driven by Fire

Large trees disproportionately contribute to biomass storage, habitat structure, and ecosystem functioning. However, their distribution and health dynamics remain poorly quantified at a regional scale. Here, a deep learning model (U-Net-ID) and canopy height models derived from sub-meter aerial imagery from 2020 were used to delineate all individual trees with crown area $\geq$ 100 m$^2$ across the Sierra Nevada Floristic Province. The model was trained using more than 3.3 million synthetic tree crowns and achieved a median Intersection over Union (IoU) of 0.602 when validated against an independent dataset of 20,273 crowns. A total of 6,515,705 large trees were mapped, occurring across approximately 78.7% of the Sierra Nevada Floristic Province. The spatial distribution of large trees showed associations with elevation, temperature, and precipitation. Using Sentinel-2 time series from 2020 to 2025, tree health dynamics were characterized by extracting spectral trajectories for each crown and applying BFAST breakpoint detection algorithm combined with a disturbance classification framework to identify mortality, disturbance, and recovery trajectories of individual trees. Wildfires, estimated from CAL FIRE fire perimeters, were identified as the dominant driver of large-tree mortality, killing 10% of all large trees in the Sierra Nevada, with mortality strongly concentrated during the extreme 2020-2021 fire seasons.

cs.CV↗