arXiv subjects

Hao Yang

Publications and source records attributed to Hao Yang.

At least 55 records · Page 3Linked to original sources

arXiv2026

Noncoherent ISAC over Block-Fading Channels: Asymptotic Performance Analysis

This paper investigates the fundamental limits and optimal signal distribution design for Integrated Sensing and Communication (ISAC) systems operating under strictly noncoherent conditions. Unlike conventional coherent frameworks that rely on perfect channel state information, we consider a block-fading MIMO channel where the channel realizations are unknown to both the transmitter and the receiver. We adopt a realization-wise perspective to characterize the noncoherent performance tradeoff across different signal-to-noise ratio (SNR) regimes. In the high-SNR regime, we derive a lower bound for the noncoherent mutual information and define a metric, termed sensing-induced rate loss, to quantify the communication penalty incurred by sensing-oriented beamforming. We then employ a projected gradient algorithm to optimize the spatial power allocation, balancing the conflict between the unitary space-time modulation-based structure for communication and the task-oriented spatial power allocation for sensing. Conversely, in the low-SNR regime, we perform a first-order asymptotic analysis of the ergodic minimum mean squared error (EMMSE). Our theoretical derivation reveals a fundamental synergy: the sensing-optimal strategy collapses to a rank-one transmission along the dominant eigenvector of the target response, which incurs no first-order communication loss in the low-SNR regime. This result demonstrates that the conflicting tradeoff observed at high SNR vanishes asymptotically at low SNR, enabling perfect alignment between sensing and communication objectives.

cs.IT↗

arXiv2026

Reasoning for Mobile User Experience with Multimodal LLMs: Task, Benchmark, and Approach

User experience (UX) centered on usability, perceived consistency, and functional clarity is fundamental to real-world user interfaces (UI). The application of multimodal large language models (MLLMs) in the field of user interfaces is evolving rapidly, such as visual element grounding, graphical user interface (GUI) agents, and design-to-code generation. However, research efforts on evaluating UX based on UI screenshots are still immature. To address this, we propose UXBench, a novel multimodal benchmark consisting of 2,000 VQA data samples designed to assess MLLMs' ability to perform UI-based reasoning. UXBench includes 8 tasks based on real-world UI screenshots that require fine-grained diagnosis of UX issues across layout relationships, visual hierarchy, and content consistency. Our extensive evaluation of mainstream MLLMs shows that they remain fundamentally limited in their capacity for UI-based reasoning. The results underscore the need for further advancements in this area. To bridge this gap, we propose UI-UX, an MLLM based on Qwen3-VL-4B-Thinking foundation model and enhanced via reinforcement learning with two key innovations: a reward routing mechanism that dynamically balances perceptual understanding and logical reasoning during inference, and an asymmetric transition reward that suppresses redundant or insufficient reasoning steps. Experiments demonstrate that UI-UX achieves state-of-the-art (SOTA) performance on UXBench, attaining an accuracy of 0.7963 -- surpassing Claude-4.5-Sonnet's 0.6550 -- while exhibiting strong generalization across diverse UI tasks and maintaining low inference latency.

cs.AI↗

arXiv2026

Rosetta Memory: Adaptive Memory for Cross-LLM Agents

Memory is the key component for transforming a stateless LLM into a persistent, evolving agent through experience accumulation, long-horizon planning, and continual self-improvement. Existing memory systems typically take the LLM as the center and design memory operations tailored to a specific backbone. In practice, however, users frequently switch between LLMs, for example using Claude for coding and GPT for writing across tasks, or routing different steps to different backbones within a single task for cost-effective trade-offs. As a result, memory written by one model often needs to be consumed by another. Making upstream memory effectively adapt to and activate downstream LLMs remains a critical yet underexplored problem. To bridge this gap, we shift the perspective from LLM-centric memory design to \emph{memory-centric LLM adaptation}. Specifically, we approach the above upstream-downstream memory adaptation problem from both the write and read sides, and design two profile-conditioned operators that are jointly trained to optimize how memory is stored and presented for better task completion. To ensure the learned operators generalize across a broad set of LLMs, we propose a minimum-gain sampling curriculum that prioritizes the least-served LLMs during training. To better measure the operators' actual contribution rather than the LLM's own capability, we design a performance-gap reward that compares against a naive memory baseline. Experiments on HotpotQA, 2WikiMultihopQA, and MuSiQue demonstrate that our model consistently outperforms baselines and remains robust under unseen-model replacement.

cs.LG↗

arXiv2026

Veda: Scalable Video Diffusion via Distilled Sparse Attention

Scaling Diffusion Transformers to generate high-resolution, long videos is constrained by the quadratic cost of self-attention, and existing sparse attention methods degrade under high sparsity. We show empirically that generation quality is determined not by the sparsity ratio itself, but by how well the sparse mask aligns with the tile-wise geometry of full attention. Based on this insight, we propose Veda, a distilled sparse attention framework that formulates tile selection as an explicit reconstruction problem from full attention. Veda integrates statistics-aware tile scoring with head-aware tiling to reduce estimation error and structural mismatch, enabling aggressive sparsity. A hardware-efficient tile-skipping kernel converts theoretical sparsity into practical wall-clock speedups. Experiments on large video diffusion models, including Waver and Wan2.1, demonstrate substantial acceleration with no noticeable degradation in generation quality. To generate 720P 10-second videos on Waver-T2V-12B, Veda achieves a 5.1$\times$ end-to-end speedup and a 10.5$\times$ self-attention speedup, reducing attention overhead from 92% to 50%. Notably, the gains increase with sequence length, indicating that Veda scales favorably with spatiotemporal resolution across models.

cs.CV↗

arXiv2026

Finite Curvature Construction of Regular Black Holes and Quasinormal Mode Analysis

We develop a class of regular black holes by prescribing finite curvature invariants and reconstructing the corresponding spacetime geometry. Two distinct approaches are employed: one based on the Ricci scalar and the other on the Weyl scalar. In each case, we explore a variety of analytic profiles for the curvature functions, including Gaussian, hyperbolic secant, and rational forms, ensuring regularity, asymptotic flatness, and compatibility with dominant energy conditions. The resulting mass functions yield spacetime geometries free from curvature singularities and exhibit horizons depending on model parameters. To assess the stability of these solutions, we perform a detailed analysis of quasinormal modes (QNMs) under axial gravitational perturbations. We show that the shape of the effective potential, particularly its width and the presence of potential valleys, plays a critical role in determining the QNMs. Models with a large peak-to-valley ratio in the potential barrier exhibit stable, exponentially decaying waveforms, while a small ratio may induce late-time instabilities. Our results highlight the significance of potential design in constructing physically viable and dynamically stable regular black holes, offering potential observational implications in modified gravity and quantum gravity scenarios.

gr-qc↗

arXiv2026

Multiple Superconducting Phases in Rhombohedral Heptalayer Graphene

Crystalline rhombohedral multilayer graphene (RMG) has emerged as an ideal platform for studying unconventional superconductivity. Here, we report the observation of superconductivity in moiréless rhombohedral heptalayer graphene (RHG) at zero magnetic field. The superconducting phases emerge at low displacement electric fields (|D| < 0.2 V/nm) and are symmetrically distributed about D = 0, with one robust state exhibiting zero resistance and two weaker superconducting features. Comparisons with rhombohedral pentalayer graphene (RPG) reveal distinct perpendicular magnetic-field responses, and quantum oscillation measurements indicate that superconductivity in RHG arises from a half-metallic normal state. These results highlight the strong dependence of superconductivity on layer number and electronic structure in RMG systems and provide new insights into its microscopic origin.

cond-mat.mes-hall↗

arXiv2026

Intuitive Surgical SurgToolLoc and SurgVU Challenges Results: 2022-2025

Robotic assisted (RA) surgery promises to transform surgical intervention. Intuitive Surgical is committed to fostering these changes and the machine learning models and algorithms that will enable them. With these goals in mind we have invited the surgical data science community to participate in a yearly competition hosted through the Medical Imaging Computing and Computer Assisted Interventions (MICCAI) conference. With varying changes from year to year, we have challenged the community to solve difficult machine learning problems in the context of advanced RA applications. Here we document the results of these challenges, focusing on surgical tool localization (SurgToolLoc) and surgical visual understanding (SurgVU). The publicly released dataset that accompanies these challenges is detailed in a separate paper arXiv:2501.09209 [1].

cs.CV↗

arXiv2026

Text-RSIR: A Text-Guided Framework for Efficient Remote Sensing Image Transmission and Reconstruction

High-resolution remote sensing imagery is critical for environmental monitoring, urban mapping, and land cover analysis, but its transmission is often hindered by limited bandwidth and high communication costs. Conventional pipelines transmit full-resolution pixel data, resulting in redundant and inefficient delivery. This paper proposes a text-guided remote sensing image transmission system that replaces complete high-resolution data with low-resolution images accompanied by compact textual descriptions. An onboard text generator produces spatial and semantic summaries, reducing the transmitted data volume to approximately 2\% of the original size. For ground-based reconstruction, a text-conditioned image restoration model is introduced, which leverages cross-modal learning to recover fine spatial details and maintain semantic coherence. Experimental results on the Alsat-2B, UC Merced Land Use, and Aerial Image datasets demonstrate that the proposed framework achieves reconstruction PSNRs of 16.36 dB, 26.87 dB, and 27.41 dB, respectively, enabling efficient and information-preserving image transfer for remote sensing applications. The implementation will be made publicly available at \href{https://github.com/haoyangofficial/textrssr}{GitHub}.

eess.IV↗

arXiv2026

A Cross-Modal Prompt Injection Attack against Large Vision-Language Models with Image-Only Perturbation

Large vision-language models (LVLMs) have emerged as a powerful paradigm for multimodal intelligence, but their growing deployment also expands the attack surface of prompt injection. Despite this growing concern, existing attacks still suffer from a critical limitation: the injected prompt for one modality only steers the model's interpretation of that singular input. Alternatively, these attacks remain multimodal but fail to achieve cross-modal prompt perturbation. To bridge this gap, we introduce a novel cross-modal prompt injection attack CrossMPI, which can steer the model's interpretation of both textual and visual inputs via image-only prompt injection. Our design is underpinned by the following key breakthroughs. First, we turn the focus of the injected prompt perturbation optimization from the visual embedding space (typically with only $10^5$ parameters) to the model hidden state space (for multimodal information integration and with $10^7$ parameters). Then, two strategies are adopted to mitigate the optimization challenges posed by the larger parameter space. To constrain the optimized model parameter space, we introduce a layer selection strategy that identifies the layers most critical to multimodal integration. Interestingly, deviating from the past experience, our analysis reveals that the optimal layers for LVLM prompt perturbation reside in the middle of the model rather than the last. To constrain the image perturbation space, we propose a new distance-decremental perturbation budget assignment strategy that allocates budgets decrementally as the pixel distance to semantic-critical regions increases. Extensive experiments across multiple LVLMs and datasets show that our method significantly outperforms baseline approaches.

cs.CR↗

arXiv2026

Low Light Image Enhancement Challenge at NTIRE 2026

This paper presents a comprehensive review of the NTIRE 2026 Low Light Image Enhancement Challenge, highlighting the proposed solutions and final results. The objective of this challenge is to identify effective networks capable of producing clearer and visually compelling images in diverse and challenging conditions by learning representative visual cues with the purpose of restoring information loss due to low-contrast and noisy images. A total of 195 participants registered for the first track and 153 for the second track of the competition, and 22 teams ultimately submitted valid entries. This paper thoroughly evaluates the state-of-the-art advances in (joint denoising and) low-light image enhancement, showcasing the significant progress in the field, while leveraging samples of our novel dataset.

cs.CV↗

arXiv2026

NTIRE 2026 The Second Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Results

This paper presents an overview of the NTIRE 2026 Second Challenge on Day and Night Raindrop Removal for Dual-Focused Images. Building upon the success of the first edition, this challenge attracted a wide range of impressive solutions, all developed and evaluated on our real-world Raindrop Clarity dataset~\cite{jin2024raindrop}. For this edition, we adjust the dataset with 14,139 images for training, 407 images for validation, and 593 images for testing. The primary goal of this challenge is to establish a strong and practical benchmark for the removal of raindrops under various illumination and focus conditions. In total, 168 teams have registered for the competition, and 17 teams submitted valid final solutions and fact sheets for the testing phase. The submitted methods achieved strong performance on the Raindrop Clarity dataset, demonstrating the growing progress in this challenging task.

cs.CV↗

arXiv2026

ViSRA: A Video-based Spatial Reasoning Agent for Multi-modal Large Language Models

Recent advances in Multi-modal Large Language Models (MLLMs) target 3D spatial intelligence, yet the progress has been largely driven by post-training on curated benchmarks, leaving the inference-time approach relatively underexplored. In this paper, we take a training-free perspective and introduce ViSRA, a human-aligned Video-based Spatial Reasoning Agent, as a framework to probe the spatial reasoning mechanism of MLLMs. ViSRA elicits spatial reasoning in a modular and extensible manner by leveraging explicit spatial information from expert models, enabling a plug-and-play flexible paradigm. ViSRA offers two key advantages: (1) human-aligned and transferable 3D understanding rather than task-specific overfitting; and (2) no post-training computational cost along with heavy manual curation of spatial reasoning datasets. Experimental results demonstrate consistent improvement across a set of MLLMs on both existing benchmarks and unseen 3D spatial reasoning tasks, with ViSRA outperforming baselines by up to a 15.6% and 28.9% absolute margin respectively.

cs.CV↗

arXiv2026

A Hybrid Graph Neural Network for Enhanced EEG-Based Depression Detection

Graph neural networks (GNNs) are becoming increasingly popular for EEG-based depression detection. However, previous GNN-based methods fail to sufficiently consider the characteristics of depression, thus limiting their performance. Firstly, studies in neuroscience indicate that depression patients exhibit both common and individualized brain abnormal patterns. Previous GNN-based approaches typically focus either on fixed graph connections to capture common abnormal brain patterns or on adaptive connections to capture individualized patterns, which is inadequate for depression detection. Secondly, brain network exhibits a hierarchical structure, which includes the arrangement from channel-level graph to region-level graph. This hierarchical structure varies among individuals and contains significant information relevant to detecting depression. Nonetheless, previous GNN-based methods overlook these individualized hierarchical information. To address these issues, we propose a Hybrid GNN (HGNN) that merges a Common Graph Neural Network (CGNN) branch utilizing fixed connection and an Individualized Graph Neural Network (IGNN) branch employing adaptive connections. The two branches capture common and individualized depression patterns respectively, complementing each other. Furthermore, we enhance the IGNN branch with a Graph Pooling and Unpooling Module (GPUM) to extract individualized hierarchical information. Extensive experiments on two public datasets show that our model achieves state-of-the-art performance.

eess.SP↗

arXiv2026

Judge, Then Drive: A Critic-Centric Vision Language Action Framework for Autonomous Driving

Recent advances in vision language action (VLA) models have shown remarkable potential for autonomous driving by directly mapping multimodal inputs to control signals. However, previous VLA-based methods have not explicitly exploited the critic capability of VLAs to refine driving decisions, even though such capability has been well demonstrated in other LLM-based domains, thereby limiting their performance in complex closed-loop scenarios. In this work, we present a theoretically inspired two-stage framework, CriticVLA, which extends the role of VLAs from acting to judging. CriticVLA first generates a rough trajectory and then refines it through multimodal evaluation and single-step optimization guided by a VLA-based critic, yielding higher-quality driving behaviors. To support this process, we construct a large-scale synthetic dataset of 12.9 million annotated trajectories covering diverse driving scenarios, which enhances the critic's reasoning and refinement abilities. Extensive closed-loop experiments on the Bench2Drive benchmark show that CriticVLA significantly surpasses state-of-the-art baselines, achieving a 73.33% total success rate and delivering about 30% improvement in challenging scenarios.

cs.CV↗

arXiv2026

Tunable high-Chern-number Chern insulators in rhombohedral tetralayer graphene/hBN moiré superlattices

Moiré superlattices based on rhombohedral multilayer graphene have emerged as a highly tunable platform for engineering correlated topological phases. Here, we systematically investigate the transport properties of the hole-doped side in rhombohedral tetralayer graphene/ hexagonal boron nitride (hBN) moiré superlattices across a range of twist angles and alignment orientations. Notably, we observed multiple high-Chern-number Chern insulators, including the previously reported integer Chern insulator with Chern number C = -4 at moiré filling factor v = -1 and newly discovered symmetry-broken Chern insulating states with C = +3, $\pm$2, $\pm$1 at fractional moiré fillings of v = -2.5 or -2.6. These Chern insulating states emerge in both hBN alignment, but exhibit a sensitive moiré wavelength dependence. Our findings demonstrate the exceptional tunability of these high-Chern-number states via moiré wavelength, displacement electric field and external magnetic field, underscoring the distinct topological landscape realized in hole-doped RTG/hBN moiré superlattices.

cond-mat.mes-hall↗

arXiv2026

Robot Planning and Situation Handling with Active Perception

Current robots are capable of computing plans to accomplish complex tasks. However, real-world environments are inherently open and dynamic, and unforeseen situations frequently arise during plan execution, such as jamming doors and fallen objects on the floor. These situations may result from the robot's own action failures or from external disturbances, such as human activities. Detecting and handling such execution - time situations remains a significant challenge, limiting those robots' ability to achieve long-term autonomy. In this paper, we develop a planning and situation-handling framework, called VAP-TAMP, that enables robots to actively perceive and address unforeseen situations during plan execution. VAP-TAMP leverages action knowledge to strategically prompt vision-language models for active view selection and situation assessment, while constructing and reasoning over scene graphs for integrated task and motion planning. We evaluated VAP-TAMP using service tasks in simulation and on a mobile manipulation platform.

cs.RO↗

arXiv2026

Systematic analysis of $D_{(s)}$ meson semi-leptonic decays in the covariant light-front quark model

The weak decays of the $D_{(s)}$ meson provide a pivotal platform to advance our understanding of the Standard Model (SM) and to explore New Physics (NP). In recent years, experiments have collected a significant amount of data on the $D_{(s)}$ meson decays, particularly from BESIII, which provides substantial support for theoretical research. In this work, we systematically investigate the semi-leptonic decays of $D_{(s)}$ meson to pseudoscalar (P), scalar (S), vector (V), and axial-vector (A) mesons within the framework of the covariant light-front quark model (CLFQM). We calculate the form factors of the transitions $D_{(s)}\to P,S,V,A$ and the branching ratios of the corresponding semi-leptonic decays, then compare them with experimental data and results from other theoretical models. The form factor values of the transitions $D_{(s)}\to P, V$ obtained from the CLFQM are consistent with those of other theoretical models and available data in most cases. However, significant discrepancies are found in some specific $D_{(s)}\to S,A$ transitions, such as $D\to a_0(980), a_0(1450)$ and $D_{(s)} \to K_{1B}$, compared to other theoretical calculations. The predicted branching ratios for the semi-leptonic decays $D\to P(V)\ellν_\ell$ with $\ell=e,μ$ agree well with experimental data and other theoretical results in most decay channels, validating the reliability of the model. However, for some $D\to S(V)\ellν_\ell$ decays, tension exists among different theoretical predictions and experimental results. Further clarification of such differences is necessary. Our study provides important insights into the internal structures for some scalar and axial-vector mesons and serves as a theoretical reference for future experiments.

hep-ph↗

arXiv2026

Visual Reasoning through Tool-supervised Reinforcement Learning

In this paper, we investigate the problem of how to effectively master tool-use to solve complex visual reasoning tasks for Multimodal Large Language Models. To achieve that, we propose a novel Tool-supervised Reinforcement Learning (ToolsRL) framework, with direct tool supervision for more effective tool-use learning. We focus on a series of simple, native, and interpretable visual tools, including zoom-in, rotate, flip, and draw point/line, whose tool supervision is easy to collect. A reinforcement learning curriculum is developed, where the first stage is solely optimized by a set of well motivated tool-specific rewards, and the second stage is trained with the accuracy targeted rewards while allowing calling tools. In this way, tool calling capability is mastered before using tools to complete visual reasoning tasks, avoiding the potential optimization conflict among those heterogeneous tasks. Our experiments have shown that the tool-supervised curriculum training is efficient and ToolsRL can achieve strong tool-use capabilities for complex visual reasoning tasks.

cs.CV↗