arXiv subjects

Yao Yao

Publications and source records attributed to Yao Yao.

At least 55 records · Page 3Linked to original sources

arXiv2025

A Review of Machine Learning for Cavitation Intensity Recognition in Complex Industrial Systems

Cavitation intensity recognition (CIR) is a critical technology for detecting and evaluating cavitation phenomena in hydraulic machinery, with significant implications for operational safety, performance optimization, and maintenance cost reduction in complex industrial systems. Despite substantial research progress, a comprehensive review that systematically traces the development trajectory and provides explicit guidance for future research is still lacking. To bridge this gap, this paper presents a thorough review and analysis of hundreds of publications on intelligent CIR across various types of mechanical equipment from 2002 to 2025, summarizing its technological evolution and offering insights for future development. The early stages are dominated by traditional machine learning approaches that relied on manually engineered features under the guidance of domain expert knowledge. The advent of deep learning has driven the development of end-to-end models capable of automatically extracting features from multi-source signals, thereby significantly improving recognition performance and robustness. Recently, physical informed diagnostic models have been proposed to embed domain knowledge into deep learning models, which can enhance interpretability and cross-condition generalization. In the future, transfer learning, multi-modal fusion, lightweight network architectures, and the deployment of industrial agents are expected to propel CIR technology into a new stage, addressing challenges in multi-source data acquisition, standardized evaluation, and industrial implementation. The paper aims to systematically outline the evolution of CIR technology and highlight the emerging trend of integrating deep learning with physical knowledge. This provides a significant reference for researchers and practitioners in the field of intelligent cavitation diagnosis in complex industrial systems.

eess.SP↗

arXiv2025

Growth estimates for axisymmetric Euler equations without swirl

We consider the axisymmetric Euler equations in $\mathbb{R}^3$ without swirl, and establish several upper and lower bounds for the growth of solutions. On the one hand, we obtain an upper bound $t^2$ for the radial moment $\int_{\mathbb{R}^3} rω^θdx$, which is the conjectured optimal rate by Childress (Phys. D 237(14-17):1921-1925, 2008). On the other hand, for all initial data satisfying certain symmetry and sign conditions, we prove that the radial moment grows at least like $t/\log t$ as time goes to infinity, and $\|ω(\cdot,t)\|_{L^p(\mathbb{R}^3)}$ exhibits at least $t^{1/4}$ growth in the limsup sense for all $1\leq p\leq \infty$. To the best of our knowledge, this is the first result to establish power-law $L^p$-norm growth for smooth, compactly supported initial vorticity in $\mathbb{R}^3$. For these initial data, we also show that nearly all vorticity must eventually escape to $r\to\infty$ in the time-integral sense.

math.AP↗

arXiv2025

LiteVGGT: Boosting Vanilla VGGT via Geometry-aware Cached Token Merging

3D vision foundation models like Visual Geometry Grounded Transformer (VGGT) have advanced greatly in geometric perception. However, it is time-consuming and memory-intensive for long sequences, limiting application to large-scale scenes beyond hundreds of images. To address this, we propose LiteVGGT, achieving up to 10x speedup and substantial memory reduction, enabling efficient processing of 1000-image scenes. We derive two key insights for 3D reconstruction: (1) tokens from local image regions have inherent geometric correlations, leading to high similarity and computational redundancy; (2) token similarity across adjacent network layers remains stable, allowing for reusable merge decisions. Guided by these, we design a simple yet efficient strategy, dubbed geometry-aware cached token merging. We analyze each token's geometric importance, optimizing anchor token selection to better preserve key information for reconstruction. We also cache and reuse merge indices across layers, substantially reducing latency with minimal accuracy impact. This strategy retains VGGT's core performance, enabling efficient fine-tuning and FP8 quantization for further gains. Extensive experiments validate LiteVGGT's effectiveness, scalability, and robustness. Project page: https://garlicba.github.io/LiteVGGT/

cs.CV↗

arXiv2025

TEXTRIX: Latent Attribute Grid for Native Texture Generation and Beyond

Prevailing 3D texture generation methods, which often rely on multi-view fusion, are frequently hindered by inter-view inconsistencies and incomplete coverage of complex surfaces, limiting the fidelity and completeness of the generated content. To overcome these challenges, we introduce TEXTRIX, a native 3D attribute generation framework for high-fidelity texture synthesis and downstream applications such as precise 3D part segmentation. Our approach constructs a latent 3D attribute grid and leverages a Diffusion Transformer equipped with sparse attention, enabling direct coloring of 3D models in volumetric space and fundamentally avoiding the limitations of multi-view fusion. Built upon this native representation, the framework naturally extends to high-precision 3D segmentation by training the same architecture to predict semantic attributes on the grid. Extensive experiments demonstrate state-of-the-art performance on both tasks, producing seamless, high-fidelity textures and accurate 3D part segmentation with precise boundaries.

cs.CV↗

arXiv2025

Hallo4: High-Fidelity Dynamic Portrait Animation via Direct Preference Optimization

Generating highly dynamic and photorealistic portrait animations driven by audio and skeletal motion remains challenging due to the need for precise lip synchronization, natural facial expressions, and high-fidelity body motion dynamics. We propose a human-preference-aligned diffusion framework that addresses these challenges through two key innovations. First, we introduce direct preference optimization tailored for human-centric animation, leveraging a curated dataset of human preferences to align generated outputs with perceptual metrics for portrait motion-video alignment and naturalness of expression. Second, the proposed temporal motion modulation resolves spatiotemporal resolution mismatches by reshaping motion conditions into dimensionally aligned latent features through temporal channel redistribution and proportional feature expansion, preserving the fidelity of high-frequency motion details in diffusion-based synthesis. The proposed mechanism is complementary to existing UNet and DiT-based portrait diffusion approaches, and experiments demonstrate obvious improvements in lip-audio synchronization, expression vividness, body motion coherence over baseline methods, alongside notable gains in human preference metrics. Our model and source code can be found at: https://github.com/fudan-generative-vision/hallo4.

cs.CV↗

arXiv2025

Advantage of utilizing nonlocal magic resource in Haar-random circuits

Magic resources and entanglement are fundamental components for achieving the universal quantum computation, so is the interplay between them. Herein, we uncover an intrinsic scaling law of the magic resource and bond dimension of matrix product states in Haar-random quantum circuits, that is, the magic resource is converged on a bond dimension in logarithmic scale with the system size. From a practical perspective, this finding substantially enhances the classical simulability of nonstabilizerness. It also allows us to utilize the bond dimension as a bridge to link the entanglement and the nonlocal magic resource, which extends the capacity perspective that the entanglement plays the role of container for the nonlocal magic resource. Furthermore, the intrinsic scaling enables an information separation between the nonlocal magic resource and the extra entanglement. This, in turn, leads to the conclusion that, any dynamical relation between magic and entanglement resources is ruled out. In other words, it is inappropriate to regard the entanglement as the driving force of the growth and spreading of nonlocal magic resource.

quant-ph↗

arXiv2025

UniArt: Unified 3D Representation for Generating 3D Articulated Objects with Open-Set Articulation

Articulated 3D objects play a vital role in realistic simulation and embodied robotics, yet manually constructing such assets remains costly and difficult to scale. In this paper, we present UniArt, a diffusion-based framework that directly synthesizes fully articulated 3D objects from a single image in an end-to-end manner. Unlike prior multi-stage techniques, UniArt establishes a unified latent representation that jointly encodes geometry, texture, part segmentation, and kinematic parameters. We introduce a reversible joint-to-voxel embedding, which spatially aligns articulation features with volumetric geometry, enabling the model to learn coherent motion behaviors alongside structural formation. Furthermore, we formulate articulation type prediction as an open-set problem, removing the need for fixed joint semantics and allowing generalization to novel joint categories and unseen object types. Experiments on the PartNet-Mobility benchmark demonstrate that UniArt achieves state-of-the-art mesh quality and articulation accuracy.

cs.CV↗

arXiv2025

Pressure2Motion: Hierarchical Human Motion Reconstruction from Ground Pressure with Text Guidance

We present Pressure2Motion, a novel motion capture algorithm that reconstructs human motion from a ground pressure sequence and text prompt. At inference time, Pressure2Motion requires only a pressure mat, eliminating the need for specialized lighting setups, cameras, or wearable devices, making it suitable for privacy-preserving, low-light, and low-cost motion capture scenarios. Such a task is severely ill-posed due to the indeterminacy of pressure signals with respect to full-body motion. To address this issue, we introduce Pressure2Motion, a generative model that leverages pressure features as input and utilizes a text prompt as a high-level guiding constraint to resolve ambiguities. Specifically, our model adopts a dual-level feature extractor to accurately interpret pressure data, followed by a hierarchical diffusion model that discerns broad-scale movement trajectories and subtle posture adjustments. Both the physical cues gained from the pressure sequence and the semantic guidance derived from descriptive texts are leveraged to guide the motion estimation with precision. To the best of our knowledge, Pressure2Motion is a pioneering work in leveraging both pressure data and linguistic priors for motion reconstruction, and the established MPL benchmark is the first benchmark for this novel motion capture task. Experiments show that our method generates high-fidelity, physically plausible motions, establishing a new state of the art for this task. The codes and benchmarks will be publicly released upon publication.

cs.CV↗

arXiv2025

From Drone Imagery to Livability Mapping: AI-powered Environment Perception in Rural China

The high cost of acquiring rural street view images has constrained comprehensive environmental perception in rural areas. Drone photographs, with their advantages of easy acquisition, broad coverage, and high spatial resolution, offer a viable approach for large-scale rural environmental perception. However, a systematic methodology for identifying key environmental elements from drone photographs and quantifying their impact on environmental perception remains lacking. To address this gap, a Vision-Language Contrastive Ranking Framework (VLCR) is designed for rural livability assessment in China. The framework employs chain-of-thought prompting strategies to guide multimodal large language models (MLLMs) in identifying visual features related to quality of life and ecological habitability from drone photographs. Subsequently, to address the instability in pairwise village comparison, a text description-constrained drone photograph comparison strategy is proposed. Finally, to overcome the efficiency bottleneck in nationwide pairwise village comparisons, an innovation ranking algorithm based on binary search interpolation is developed, which reduces the number of comparisons through automated selection of comparison targets. The proposed framework achieves superior performance with a Spearman Footrule distance of 0.74, outperforming mainstream commercial MLLMs by approximately 0.1. Moreover, the mechanism of concurrent comparison and ranking demonstrates a threefold enhancement in computational efficiency. Our framework has achieved data innovation and methodological breakthroughs in village livability assessment, providing strong support for large-scale village livability analysis. Keywords: Drone photographs, Environmental perception, Rural livability assessment, Multimodal large language models, Chain-of-thought prompting.

cs.CY↗

arXiv2025

XQuant: Achieving Ultra-Low Bit KV Cache Quantization with Cross-Layer Compression

Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse natural language processing tasks. However, their extensive memory requirements, particularly due to KV cache growth during long-text understanding and generation, present significant challenges for deployment in resource-constrained environments. Quantization has emerged as a promising solution to reduce memory consumption while preserving historical information. We propose XQuant, a training-free and plug-and-play framework that achieves ultra-low equivalent bit-width KV cache quantization. XQuant introduces two key innovations: a computationally negligible data-free calibration method and cross-layer KV cache compression, enabling quantization to sub-1.4 bits. Extensive experiments on TruthfulQA and LongBench demonstrate that XQuant outperforms state-of-the-art methods (e.g., KIVI-2bit and AsymKV-1.5bit) by achieving lower bit-width while maintaining superior performance, establishing a better trade-off between memory efficiency and model accuracy.

cs.CL↗

arXiv2025

ComGS: Efficient 3D Object-Scene Composition via Surface Octahedral Probes

Gaussian Splatting (GS) enables immersive rendering, but realistic 3D object-scene composition remains challenging. Baked appearance and shadow information in GS radiance fields cause inconsistencies when combining objects and scenes. Addressing this requires relightable object reconstruction and scene lighting estimation. For relightable object reconstruction, existing Gaussian-based inverse rendering methods often rely on ray tracing, leading to low efficiency. We introduce Surface Octahedral Probes (SOPs), which store lighting and occlusion information and allow efficient 3D querying via interpolation, avoiding expensive ray tracing. SOPs provide at least a 2x speedup in reconstruction and enable real-time shadow computation in Gaussian scenes. For lighting estimation, existing Gaussian-based inverse rendering methods struggle to model intricate light transport and often fail in complex scenes, while learning-based methods predict lighting from a single image and are viewpoint-sensitive. We observe that 3D object-scene composition primarily concerns the object's appearance and nearby shadows. Thus, we simplify the challenging task of full scene lighting estimation by focusing on the environment lighting at the object's placement. Specifically, we capture a 360 degrees reconstructed radiance field of the scene at the location and fine-tune a diffusion model to complete the lighting. Building on these advances, we propose ComGS, a novel 3D object-scene composition framework. Our method achieves high-quality, real-time rendering at around 28 FPS, produces visually harmonious results with vivid shadows, and requires only 36 seconds for editing. Code and dataset are available at https://nju-3dv.github.io/projects/ComGS/.

cs.CV↗

arXiv2025

Balanced ternary formalism of second quantization

We construct a second-quantized representation with a structure of balanced ternary formalism, which involves three substances in organic molecular materials, namely electron, hole and charge-transfer exciton, into a uniform framework. The quantum thermodynamic of excitons is investigated in a closed and compact manner, benefitting from the interplay of the three substances. In order to be friendly with quantum simulations, the interactions among them are all described with unitary transformations. Significantly, the nonconserving dynamics of particle numbers, such as the generation of charge current and the exciton fission in organic semiconductors, is consistently expressed by this unitary formalism on the basis of bosonic coherent states. The spin degree of freedom is further taken into account, and an exotic molecular ferromagnetic ordering is induced in a specific configuration of excitons. This balanced ternary formalism establishes a solid bridge to connect thermodynamics and quantum simulations.

quant-ph↗

arXiv2025

Caution for the Environment: Multimodal LLM Agents are Susceptible to Environmental Distractions

This paper investigates the faithfulness of multimodal large language model (MLLM) agents in a graphical user interface (GUI) environment, aiming to address the research question of whether multimodal GUI agents can be distracted by environmental context. A general scenario is proposed where both the user and the agent are benign, and the environment, while not malicious, contains unrelated content. A wide range of MLLMs are evaluated as GUI agents using a simulated dataset, following three working patterns with different levels of perception. Experimental results reveal that even the most powerful models, whether generalist agents or specialist GUI agents, are susceptible to distractions. While recent studies predominantly focus on the helpfulness of agents, our findings first indicate that these agents are prone to environmental distractions. Furthermore, we implement an adversarial environment injection and analyze the approach to improve faithfulness, calling for a collective focus on this important topic.

cs.CL↗

arXiv2025

A Survey: Learning Embodied Intelligence from Physical Simulators and World Models

The pursuit of artificial general intelligence (AGI) has placed embodied intelligence at the forefront of robotics research. Embodied intelligence focuses on agents capable of perceiving, reasoning, and acting within the physical world. Achieving robust embodied intelligence requires not only advanced perception and control, but also the ability to ground abstract cognition in real-world interactions. Two foundational technologies, physical simulators and world models, have emerged as critical enablers in this quest. Physical simulators provide controlled, high-fidelity environments for training and evaluating robotic agents, allowing safe and efficient development of complex behaviors. In contrast, world models empower robots with internal representations of their surroundings, enabling predictive planning and adaptive decision-making beyond direct sensory input. This survey systematically reviews recent advances in learning embodied AI through the integration of physical simulators and world models. We analyze their complementary roles in enhancing autonomy, adaptability, and generalization in intelligent robots, and discuss the interplay between external simulation and internal modeling in bridging the gap between simulated training and real-world deployment. By synthesizing current progress and identifying open challenges, this survey aims to provide a comprehensive perspective on the path toward more capable and generalizable embodied AI systems. We also maintain an active repository that contains up-to-date literature and open-source projects at https://github.com/NJU3DV-LoongGroup/Embodied-World-Models-Survey.

cs.RO↗

arXiv2025

Ultranarrow Superradiant Laser via Dressed Dark State

Superradiant laser, which exploits the clock transition of alkaline-earth-metal-like atoms to generate ultrastable light in the bad-cavity limit, has garnered much attention in the past few decades. Unlike their odd counterpart, the even isotopes of alkaline-earth-metal-like atoms possess simpler structures and longer-lived transitions, which would relax the field control requirements and enhance the frequency stability of the output light. However, due to the absence of hyper-fine interaction in even isotopes, the transition from the state $^{3}\mathrm{P}_{0}$ to $^{1}$S$_{0}$ is strictly forbidden, leading to a vanishing coupling strength between the cavity mode and the atoms (the state $^{3}\mathrm{P}_{0}$ as a dark state). In this work, we suggest a superradiant laser scheme by dressing this dark state with a small bright component by virtue of a static magnetic field. In contrast to other proposals utilizing natural atomic transitions, our dressed-state protocol can work from the crossover regime (coherence in both atoms and photons) to the superradiant lasing regime (coherence solely in atoms). Specially, by operating deep into the superradiant lasing regime, our scheme witnesses a dramatic line-narrowing feature (mHz level) while maintaining its power. Furthermore, compared to the crossover regime, the laser frequency in the superradiant lasing regime is more robust against the fluctuations of the cavity length and magnetic field strength. Our proposal demonstrates the potential for extracting ultranarrow light from even isotopes of the alkaline-earth-metal-like atoms and may find its role in frequency stabilization and related precision measurement scenarios.

physics.atom-ph↗

arXiv2025

Reaction enhancement by flux-limited chemotaxis

Chemotaxis plays a crucial role in a variety of processes in biology and ecology. Quite often it acts to improve efficiency of biological reactions; one example is the immune system signalling, where infected tissues release chemokines attracting monocytes to fight invading bacteria. Another example is reproduction, where eggs release pheromones that attract sperm. In this paper, we analyze a system of two reacting densities, one of which is chemotactic on another. Since the speed of any biological agents is limited, we employ flux limited chemotaxis model. Our main result is the rigorous derivation of the scaling laws showing how presence of chemotaxis affects the typical reaction time scale. This work builds on the results of \cite{kiselev2022chemotaxis}, which employed a classical Keller-Segel chemotaxis term (not flux limited) - leading to the effect of possible over concentration and restricting the results to radial data. The model presented here is more reasonable biologically and covers broader parameter regimes.

math.AP↗

arXiv2025

Can Multi-modal (reasoning) LLMs detect document manipulation?

Document fraud poses a significant threat to industries reliant on secure and verifiable documentation, necessitating robust detection mechanisms. This study investigates the efficacy of state-of-the-art multi-modal large language models (LLMs)-including OpenAI O1, OpenAI 4o, Gemini Flash (thinking), Deepseek Janus, Grok, Llama 3.2 and 4, Qwen 2 and 2.5 VL, Mistral Pixtral, and Claude 3.5 and 3.7 Sonnet-in detecting fraudulent documents. We benchmark these models against each other and prior work on document fraud detection techniques using a standard dataset with real transactional documents. Through prompt optimization and detailed analysis of the models' reasoning processes, we evaluate their ability to identify subtle indicators of fraud, such as tampered text, misaligned formatting, and inconsistent transactional sums. Our results reveal that top-performing multi-modal LLMs demonstrate superior zero-shot generalization, outperforming conventional methods on out-of-distribution datasets, while several vision LLMs exhibit inconsistent or subpar performance. Notably, model size and advanced reasoning capabilities show limited correlation with detection accuracy, suggesting task-specific fine-tuning is critical. This study underscores the potential of multi-modal LLMs in enhancing document fraud detection systems and provides a foundation for future research into interpretable and scalable fraud mitigation strategies.

cs.CV↗

arXiv2025

Superlinear gradient growth for 2D Euler equation without boundary

We consider the vorticity gradient growth of solutions to the two-dimensional Euler equations in domains without boundary, namely in the torus $\mathbb{T}^{2}$ and the whole plane $\mathbb{R}^{2}$. In the torus, whenever we have a steady state $ω^*$ that is orbitally stable up to a translation and has a saddle point, we construct ${\tildeω}_0 \in C^\infty(\mathbb{T}^2)$ that is arbitrarily close to $ω^*$ in $L^2$, such that superlinear growth of the vorticity gradient occurs for an open set of smooth initial data around ${\tildeω}_0$. This seems to be the first superlinear growth result which holds for an open set of smooth initial data (and does not require any symmetry assumptions on the initial vorticity). Furthermore, we obtain the first superlinear growth result for smooth and compactly supported vorticity in the plane, using perturbations of the Lamb-Chaplygin dipole.

math.AP↗