arXiv subjects

Hao Wu

Publications and source records attributed to Hao Wu.

At least 307 records · Page 17Linked to original sources

arXiv2025

KineDiff3D: Kinematic-Aware Diffusion for Category-Level Articulated Object Shape Reconstruction and Generation

Articulated objects, such as laptops and drawers, exhibit significant challenges for 3D reconstruction and pose estimation due to their multi-part geometries and variable joint configurations, which introduce structural diversity across different states. To address these challenges, we propose KineDiff3D: Kinematic-Aware Diffusion for Category-Level Articulated Object Shape Reconstruction and Generation, a unified framework for reconstructing diverse articulated instances and pose estimation from single view input. Specifically, we first encode complete geometry (SDFs), joint angles, and part segmentation into a structured latent space via a novel Kinematic-Aware VAE (KA-VAE). In addition, we employ two conditional diffusion models: one for regressing global pose (SE(3)) and joint parameters, and another for generating the kinematic-aware latent code from partial observations. Finally, we produce an iterative optimization module that bidirectionally refines reconstruction accuracy and kinematic parameters via Chamfer-distance minimization while preserving articulation constraints. Experimental results on synthetic, semi-synthetic, and real-world datasets demonstrate the effectiveness of our approach in accurately reconstructing articulated objects and estimating their kinematic properties.

cs.CV↗

arXiv2025

GI-NAS: Boosting Gradient Inversion Attacks Through Adaptive Neural Architecture Search

Gradient Inversion Attacks invert the transmitted gradients in Federated Learning (FL) systems to reconstruct the sensitive data of local clients and have raised considerable privacy concerns. A majority of gradient inversion methods rely heavily on explicit prior knowledge (e.g., a well pre-trained generative model), which is often unavailable in realistic scenarios. This is because real-world client data distributions are often highly heterogeneous, domain-specific, and unavailable to attackers, making it impractical for attackers to obtain perfectly matched pre-trained models, which inevitably suffer from fundamental distribution shifts relative to target private data. To alleviate this issue, researchers have proposed to leverage the implicit prior knowledge of an over-parameterized network. However, they only utilize a fixed neural architecture for all the attack settings. This would hinder the adaptive use of implicit architectural priors and consequently limit the generalizability. In this paper, we further exploit such implicit prior knowledge by proposing Gradient Inversion via Neural Architecture Search (GI-NAS), which adaptively searches the network and captures the implicit priors behind neural architectures. Extensive experiments verify that our proposed GI-NAS can achieve superior attack performance compared to state-of-the-art gradient inversion methods, even under more practical settings with high-resolution images, large-sized batches, and advanced defense strategies. To the best of our knowledge, we are the first to successfully introduce NAS to the gradient inversion community. We believe that this work exposes critical vulnerabilities in real-world federated learning by demonstrating high-fidelity reconstruction of sensitive data without requiring domain-specific priors, forcing urgent reassessment of FL privacy safeguards.

cs.AI↗

arXiv2025

Spatiotemporal Forecasting as Planning: A Model-Based Reinforcement Learning Approach with Generative World Models

To address the dual challenges of inherent stochasticity and non-differentiable metrics in physical spatiotemporal forecasting, we propose Spatiotemporal Forecasting as Planning (SFP), a new paradigm grounded in Model-Based Reinforcement Learning. SFP constructs a novel Generative World Model to simulate diverse, high-fidelity future states, enabling an "imagination-based" environmental simulation. Within this framework, a base forecasting model acts as an agent, guided by a beam search-based planning algorithm that leverages non-differentiable domain metrics as reward signals to explore high-return future sequences. These identified high-reward candidates then serve as pseudo-labels to continuously optimize the agent's policy through iterative self-training, significantly reducing prediction error and demonstrating exceptional performance on critical domain metrics like capturing extreme events.

cs.LG↗

arXiv2025

OneForecast: A Universal Framework for Global and Regional Weather Forecasting

Accurate weather forecasts are important for disaster prevention, agricultural planning, etc. Traditional numerical weather prediction (NWP) methods offer physically interpretable high-accuracy predictions but are computationally expensive and fail to fully leverage rapidly growing historical data. In recent years, deep learning models have made significant progress in weather forecasting, but challenges remain, such as balancing global and regional high-resolution forecasts, excessive smoothing in extreme event predictions, and insufficient dynamic system modeling. To address these issues, this paper proposes a global-regional nested weather forecasting framework (OneForecast) based on graph neural networks. By combining a dynamic system perspective with multi-grid theory, we construct a multi-scale graph structure and densify the target region to capture local high-frequency features. We introduce an adaptive messaging mechanism, using dynamic gating units to deeply integrate node and edge features for more accurate extreme event forecasting. For high-resolution regional forecasts, we propose a neural nested grid method to mitigate boundary information loss. Experimental results show that OneForecast performs excellently across global to regional scales and short-term to long-term forecasts, especially in extreme event predictions. Codes link https://github.com/YuanGao-YG/OneForecast.

cs.LG↗

arXiv2025

SaFeR-VLM: Toward Safety-aware Fine-grained Reasoning in Multimodal Models

Multimodal Large Reasoning Models (MLRMs) demonstrate impressive cross-modal reasoning but often amplify safety risks under adversarial or unsafe prompts, a phenomenon we call the \textit{Reasoning Tax}. Existing defenses mainly act at the output level and do not constrain the reasoning process, leaving models exposed to implicit risks. In this paper, we propose SaFeR-VLM, a safety-aligned reinforcement learning framework that embeds safety directly into multimodal reasoning. The framework integrates four components: (I) QI-Safe-10K, a curated dataset emphasizing safety-critical and reasoning-sensitive cases; (II) safety-aware rollout, where unsafe generations undergo reflection and correction instead of being discarded; (III) structured reward modeling with multi-dimensional weighted criteria and explicit penalties for hallucinations and contradictions; and (IV) GRPO optimization, which reinforces both safe and corrected trajectories. This unified design shifts safety from a passive safeguard to an active driver of reasoning, enabling scalable and generalizable safety-aware reasoning. SaFeR-VLM further demonstrates robustness against both explicit and implicit risks, supporting dynamic and interpretable safety decisions beyond surface-level filtering. SaFeR-VLM-3B achieves average performance $70.13$ and $78.97$ on safety and helpfulness across six benchmarks, surpassing both same-scale and $>10\times$ larger models such as Skywork-R1V3-38B, Qwen2.5VL-72B, and GLM4.5V-106B. Remarkably, SaFeR-VLM-7B benefits from its increased scale to surpass GPT-5-mini and Gemini-2.5-Flash by \num{6.47} and \num{16.76} points respectively on safety metrics, achieving this improvement without any degradation in helpfulness performance. Our codes are available at https://github.com/HarveyYi/SaFeR-VLM.

cs.LG↗

arXiv2025

The low-lying light tetraquark states with quantum numbers $J^{P}=0^{+ }$, $1^{+}$ and $2^{+}$

The low-lying light tetraquark states are investigated in the non-relativistic quark model (NRQM) including the pseudoscalar meson exchange, where two different confinement potential schemes, the Cornell potential and the linear potential, are employed, along with the instanton-induced interaction serving as the residual spin-dependent interaction. The numerical results show agreement with masses of $f_{0}(500)$, $f_{0}(1370)$, $f_{0}(1500)$, $f_{0}(2020)$, $f_{0}(2200)$, $h_{1}(1170)$, $h_{1}(1595)$, $h_{1}(1900)$, $h_{1}(1965)$, $h_{1}(2215)$, $f_{2}(1430)$, $f_{2}(1640)$, $f_{2}(1810)$, $f_{2}(2010)$, $f_{2}(2150)$, $a_{0}(980)$, $a_{0}(1450)$, $a_{0}(1950)$, $a_{1}(1260)$, $a_{1}(1640)$, $a_{2}(1700)$, $K^{*}_{0}(1430)$, $K^{*}_{0}(1950)$, $K_{1}(1270)$, $K_{1}(1440)$, $K_{1}(1650)$, and $K^{*}_{2}(1980)$. The results shed light on the spectrum of these mesons and offer guidande to search for the tetraquarks in the future.

hep-ph↗

arXiv2025

Video-in-the-Loop: Span-Grounded Long Video QA with Interleaved Reasoning

We present \emph{Video-in-the-Loop} (ViTL), a two-stage long-video QA framework that preserves a fixed token budget by first \emph{localizing} question-relevant interval(s) with a low-fps skim and then \emph{answering} via span-aware reallocation of visual tokens at higher effective frame rate, emitting an interleaved output with both spans and the final option for direct attribution. We also introduce \dataname{}, which converts description based event graphs into \emph{span-grounded} multiple-choice QA by pairing each question with \emph{ground-truth} time span(s) and related reasoning. ViTL is trained end-to-end with an interleaved group-relative objective that couples temporal IoU for localization with answer correctness, allowing credit to flow from answers back to spans without increasing compute. Under fixed token budgets, ViTL attains up to 8.6% with 50% less frame input on long-video QA and temporal grounding (e.g., Charades-STA, ActivityNet-Captions) and ablations show that span-aware token reallocation consistently surpasses uniform sampling. Together, \dataname{} and ViTL provide an interpretable, compute-efficient recipe for scalable long-video QA.

cs.CV↗

arXiv2025

The Analysis and Performance of LODC-OFDM Signal in Nonlinear Rydberg Atomic Sensor

Rydberg atomic sensors have been seen as novel radio frequency (RF) measurements and the high sensitivity to a large range of frequencies makes it attractive for communications reception. However, the signal sensing process in Rydberg system involves sequential transduction from electromagnetic waves to optical signals and finally to electrical signals. The unipolar characteristic of the optical interface inherently restricts conventional OFDM reception. Therefore, adopting unipolar OFDM schemes, inspired by optical communication systems, becomes essential for compatible signal transmission. In this work, we investigate the amplitude modulation-to-amplitude modulation (AM-AM) characteristics of Rydberg atomic sensors, establishing an empirical approximation function. Building on the direct current-biased optical orthogonal frequency division multiplexing (DCO-OFDM) framework, we propose a novel local oscillator direct current-biased OFDM (LODC-OFDM) scheme specifically optimized for Rydberg-based sensing, effectively addressing the broadband OFDM reception challenge. Then, we adopt Bussgang theorem to analyze the nonlinear distortion of LODC-OFDM signals and the results in closed-form solutions are derived for AM/AM curves approximated by Taylor series expansion and for the ideal pre-distortion case. In real experiments, the experimental and theoretical results fit well.

eess.SP↗

arXiv2025

InSpire: Vision-Language-Action Models with Intrinsic Spatial Reasoning

Leveraging pretrained Vision-Language Models (VLMs) to map language instruction and visual observations to raw low-level actions, Vision-Language-Action models (VLAs) hold great promise for achieving general-purpose robotic systems. Despite their advancements, existing VLAs tend to spuriously correlate task-irrelevant visual features with actions, limiting their generalization capacity beyond the training data. To tackle this challenge, we propose Intrinsic Spatial Reasoning (InSpire), a simple yet effective approach that mitigates the adverse effects of spurious correlations by boosting the spatial reasoning ability of VLAs. Specifically, InSpire redirects the VLA's attention to task-relevant factors by prepending the question "In which direction is the [object] relative to the robot?" to the language instruction and aligning the answer "right/left/up/down/front/back/grasped" and predicted actions with ground-truth. Notably, InSpire can be used as a plugin to enhance existing autoregressive VLAs, requiring no extra training data or interaction with other large models. Extensive experimental results in both simulation and real-world environments demonstrate the effectiveness and flexibility of our approach.

cs.RO↗

arXiv2025

AdaNav: Adaptive Reasoning with Uncertainty for Vision-Language Navigation

Vision Language Navigation (VLN) requires agents to follow natural language instructions by grounding them in sequential visual observations over long horizons. Explicit reasoning could enhance temporal consistency and perception action alignment, but reasoning at fixed steps often leads to suboptimal performance and unnecessary computation. To address this, we propose AdaNav, an uncertainty-based adaptive reasoning framework for VLN. At its core is the Uncertainty Adaptive Reasoning Block (UAR), a lightweight plugin that dynamically triggers reasoning. We introduce Action Entropy as a policy prior for UAR and progressively refine it through a Heuristics to RL training method, enabling agents to learn difficulty aware reasoning policies under the strict data limitations of embodied tasks. Results show that with only 6K training samples, AdaNav achieves substantial gains over closed source models trained on million scale data, improving success rate by 20% on R2R val-unseen, 11.7% on RxR-CE, and 11.4% in real world scenes. The code is available at https://github.com/xinding-sys/AdaNav.

cs.RO↗

arXiv2025

Solvatochromic microcavity ion sensors

A novel copper(II) ion sensor has been developed, achieving both a broad dynamic range and high sensitivity, effectively addressing the limitations of conventional fluorescence-based detection methods. The sensor exploits the hydrolysis reaction between rhodamine B hydrazide (RBH) and copper(II) ions, with optical signal amplification enabled by a whispering-gallery-mode (WGM) microcavity laser. By adjusting the solvent type and utilizing the solvatochromic effect, the detection range of the copper(II) ion and RBH reaction was optimized. The sensor enables rapid, broad dynamic range detection of copper(II) ions, spanning from 5 uM to 3 mM, with a high selectivity, which holds significant potential for future applications in environmental monitoring and biomedical fields.

physics.optics↗

arXiv2025

Making Language Model a Hierarchical Classifier

Decoder-only language models, such as GPT and LLaMA, generally decode on the last layer. Motivated by human's hierarchical thinking capability, we propose that a hierarchical decoder architecture could be built with different layers decoding texts simultaneously. Due to limited time and computationally resources, we choose to adapt a pretrained language model into this form of hierarchical decoder. Language heads of the last layer are copied to different selected intermediate layers, and fine-tuned with different task inputs. By thorough experiments, we validate that these selective intermediate layers could be adapted to speak meaningful and reasonable contents, and this paradigm of hierarchical decoder can obtain state-of-the-art performances on multiple tasks such as hierarchical text classification, classification-guided generation, and hierarchical text generation. HdLM outperforms all baselines on WoS, DBpedia, ESconv, EmpatheticDialogues, and several cognitive tests. We also provide thorough theoretical analysis to validate the convergence and computational savings of our methodology. This study suggests the possibility of a generalized hierarchical reasoner, pretraining from scratch.

cs.CL↗

arXiv2025

Particle Acceleration and Transport in the Large-scale Current Sheet under an Erupting Magnetic Flux Rope

We investigate the acceleration and transport of electrons in the highly fine-structured current sheet that develops during magnetic flux rope (MFR) eruptions. Our work combines ultra-resolved MHD simulations of MFR eruption, with test-particle studies performed using the guiding center approximation. Our grid-adaptive, fully three-dimensional, high-resolution magnetohydrodynamic simulations model MFR eruptions that form complex current sheet topologies, serving as background electromagnetic fields for particle acceleration. Within the current sheet, tearing-mode instabilities give rise to mini flux ropes. Electrons become temporarily trapped within these elongated structures, undergoing acceleration and transport processes that significantly differ from those observed in two-dimensional or two-and-a-half-dimensional simulations. Our findings reveal that these fine-scale structures act as efficient particle accelerators, surpassing the acceleration efficiency of single X-line reconnection events, and are capable of energizing electrons to energies exceeding 100 keV. High-energy electrons accelerated in different mini flux ropes follow distinct trajectories due to spatially varying magnetic field connectivity, ultimately precipitating onto opposite sides of flare ribbons. Remarkably, double electron sources at the flare ribbons originate from different small flux rope acceleration regions, rather than from the same reconnecting field line as previously suggested. Distinct small flux ropes possess opposite magnetic helicity to accelerate electrons to source regions with different magnetic polarities, establishing a novel conjugate double source configuration. Furthermore, electrons escaping from the lower regions exhibit a broken power-law energy spectrum.

astro-ph.SR↗

arXiv2025

VISION: Prompting Ocean Vertical Velocity Reconstruction from Incomplete Observations

Reconstructing subsurface ocean dynamics, such as vertical velocity fields, from incomplete surface observations poses a critical challenge in Earth science, a field long hampered by the lack of standardized, analysis-ready benchmarks. To systematically address this issue and catalyze research, we first build and release KD48, a high-resolution ocean dynamics benchmark derived from petascale simulations and curated with expert-driven denoising. Building on this benchmark, we introduce VISION, a novel reconstruction paradigm based on Dynamic Prompting designed to tackle the core problem of missing data in real-world observations. The essence of VISION lies in its ability to generate a visual prompt on-the-fly from any available subset of observations, which encodes both data availability and the ocean's physical state. More importantly, we design a State-conditioned Prompting module that efficiently injects this prompt into a universal backbone, endowed with geometry- and scale-aware operators, to guide its adaptive adjustment of computational strategies. This mechanism enables VISION to precisely handle the challenges posed by varying input combinations. Extensive experiments on the KD48 benchmark demonstrate that VISION not only substantially outperforms state-of-the-art models but also exhibits strong generalization under extreme data missing scenarios. By providing a high-quality benchmark and a robust model, our work establishes a solid infrastructure for ocean science research under data uncertainty. Our codes are available at: https://github.com/YuanGao-YG/VISION.

cs.LG↗

arXiv2025

Energy based diffusion generator for efficient sampling of Boltzmann distributions

Sampling from Boltzmann distributions, particularly those tied to high dimensional and complex energy functions, poses a significant challenge in many fields. In this work, we present the Energy-Based Diffusion Generator (EDG), a novel approach that integrates ideas from variational autoencoders and diffusion models. EDG uses a decoder to generate Boltzmann-distributed samples from simple latent variables, and a diffusion-based encoder to estimate the Kullback-Leibler divergence to the target distribution. Notably, EDG is simulation-free, eliminating the need to solve ordinary or stochastic differential equations during training. Furthermore, by removing constraints such as bijectivity in the decoder, EDG allows for flexible network design. Through empirical evaluation, we demonstrate the superior performance of EDG across a variety of sampling tasks with complex target distributions, outperforming existing methods.

cs.LG↗

arXiv2025

Detection Capability Comparison Between Intensity Detection and Splitting Detection for Rydberg-Atomic Sensors

Rydberg atomic quantum receivers have been seen as novel radio frequency measurements and the high sensitivity to a large range of frequencies makes it attractive for communications reception. However, their unique physical characteristics enable two fundamental signal readout schemes: intensity-based detection and splitting-based detection. The former measures the electric fields through laser intensity, while the latter utilizes Autler-Townes splitting. In this work, we systematically categorize and model existing signal readout methods, classifying them into these two paradigms. Then, we derive the maximum likelihood estimation procedures and corresponding Cramér-Rao lower bounds (CRLB) for each detection modality. Through the analysis of the CRLB, we propose strategy for both readout schemes to enhance sensitivity and minimize estimation variance: acquiring data in regions with maximal slope magnitudes. While this approach has been implemented in intensity-based detection (e.g., superheterodyne schemes), its application to splitting-based detection remains unexplored. Implementation of non-uniform frequency scanning, with preferential sampling at regions exhibiting maximum peak slopes combined with our proposed maximum likelihood splitting estimation method, achieves significantly reduced estimation variance compared to conventional polynomial fitting. The comparative analysis reveals the optimal detection performance of the two detection schemes. This work also contributes to enhancing the accuracy of microwave calibration. Numerical results reveal that both fundamental signal readout methods achieve lower estimation variance based on our proposed maximum likelihood estimation approach.

eess.SP↗

arXiv2025

MoSEs: Uncertainty-Aware AI-Generated Text Detection via Mixture of Stylistics Experts with Conditional Thresholds

The rapid advancement of large language models has intensified public concerns about the potential misuse. Therefore, it is important to build trustworthy AI-generated text detection systems. Existing methods neglect stylistic modeling and mostly rely on static thresholds, which greatly limits the detection performance. In this paper, we propose the Mixture of Stylistic Experts (MoSEs) framework that enables stylistics-aware uncertainty quantification through conditional threshold estimation. MoSEs contain three core components, namely, the Stylistics Reference Repository (SRR), the Stylistics-Aware Router (SAR), and the Conditional Threshold Estimator (CTE). For input text, SRR can activate the appropriate reference data in SRR and provide them to CTE. Subsequently, CTE jointly models the linguistic statistical properties and semantic features to dynamically determine the optimal threshold. With a discrimination score, MoSEs yields prediction labels with the corresponding confidence level. Our framework achieves an average improvement 11.34% in detection performance compared to baselines. More inspiringly, MoSEs shows a more evident improvement 39.15% in the low-resource case. Our code is available at https://github.com/creator-xi/MoSEs.

cs.CL↗

arXiv2025

StreamMind: Unlocking Full Frame Rate Streaming Video Dialogue through Event-Gated Cognition

With the rise of real-world human-AI interaction applications, such as AI assistants, the need for Streaming Video Dialogue is critical. To address this need, we introduce StreamMind, a video LLM framework that achieves ultra-FPS streaming video processing (100 fps on a single A100) and enables proactive, always-on responses in real time, without explicit user intervention. To solve the key challenge of the contradiction between linear video streaming speed and quadratic transformer computation cost, we propose a novel perception-cognition interleaving paradigm named ''event-gated LLM invocation'', in contrast to the existing per-time-step LLM invocation. By introducing a Cognition Gate network between the video encoder and the LLM, LLM is only invoked when relevant events occur. To realize the event feature extraction with constant cost, we propose Event-Preserving Feature Extractor (EPFE) based on state-space method, generating a single perception token for spatiotemporal features. These techniques enable the video LLM with full-FPS perception and real-time cognition response. Experiments on Ego4D and SoccerNet streaming tasks, as well as standard offline benchmarks, demonstrate state-of-the-art performance in both model capability and real-time efficiency, paving the way for ultra-high-FPS applications, such as Game AI and interactive media. The code and data is available at https://aka.ms/StreamMind.

cs.CV↗