arXiv subjects

Cheng Zhang

Publications and source records attributed to Cheng Zhang.

At least 127 records · Page 7Linked to original sources

arXiv2026

Towards the Distributed Large-scale k-NN Graph Construction by Graph Merge

In order to support the real-time interaction with LLMs and the instant search or the instant recommendation on social media, it becomes an imminent problem to build a k-NN graph or an indexing graph for the massive number of vectorized multimedia data. In such scenarios, the scale of the data or the scale of the graph may exceed the processing capacity of a single machine. This paper aims to address the graph construction problem of such scale via efficient graph merge. For the graph construction on a single node, two generic and highly parallelizable algorithms, namely Two-way Merge and Multi-way Merge are proposed to merge subgraphs into one. For the graph construction across multiple nodes, a multi-node procedure based on Two-way Merge is presented. The procedure makes it feasible to construct a large-scale k-NN graph/indexing graph on either a single node or multiple nodes when the data size exceeds the memory capacity of one node. Extensive experiments are conducted on both large-scale k-NN graph and indexing graph construction. For the k-NN graph construction, the large-scale and high-quality k-NN graphs are constructed by graph merge in parallel. Typically, a billion-scale k-NN graph can be built in approximately 17h when only three nodes are employed. For the indexing graph construction, similar NN search performance as the original indexing graph is achieved with the merged indexing graphs while requiring much less time of construction.

cs.DC↗

arXiv2026

A Kernel Approach for Semi-implicit Variational Inference

Semi-implicit variational inference (SIVI) enhances the expressiveness of variational families through hierarchical semi-implicit distributions, but the intractability of their densities makes standard ELBO-based optimization biased. Recent score-matching approaches to SIVI (SIVI-SM) address this issue via a minimax formulation, at the expense of an additional lower-level optimization problem. In this paper, we propose kernel semi-implicit variational inference (KSIVI), a principled and tractable alternative that eliminates the lower-level optimization by leveraging kernel methods. We show that when optimizing over a reproducing kernel Hilbert space, the lower-level problem admits an explicit solution, reducing the objective to the kernel Stein discrepancy (KSD). Exploiting the hierarchical structure of semi-implicit distributions, the resulting KSD objective can be efficiently optimized using stochastic gradient methods. We establish optimization guarantees via variance bounds on Monte Carlo gradient estimators and derive statistical generalization bounds of order $\tilde{\mathcal{O}}(1/\sqrt{n})$. We further introduce a multi-layer hierarchical extension that improves expressiveness while preserving tractability. Empirical results on synthetic and real-world Bayesian inference tasks demonstrate the effectiveness of KSIVI.

stat.ML↗

arXiv2026

High-capacity dual degrees of freedom quantum secret sharing protocol beyond the linear rate-distance bound

Quantum secret sharing (QSS) is the multipartite cryptographic primitive. Most of existing QSS protocols are limited by the linear rate-distance bound, and cannot realize the long-distance and high-capacity multipartite key distribution. This paper proposes a polarization (Pol) and phase (Ph) dual degrees of freedom (dual-DOF) QSS protocol based on the weak coherent pulse (WCP) sources. Our protocol combines the single-photon interference, two-photon interference and non-interference principles, and can resist the internal attack from the dishonest player. We develop simulation method to estimate its performance under the beam splitting attack. The simulation results show that our protocol can surpass the linear bound. Comparing with the differential-phase-shift twin-field QSS and WCP-Ph-QSS protocols, our protocol has stronger resistance against the beam splitting attack, and thus has longer maximal communication distance and higher key rate. By using the WCPs with high average photon number ($μ$ = 1.5), our protocol achieves a key rate about 5.4 times of that in WCP-Ph-QSS protocol. Its maximal communication distance (441.7 km) is about 7.9% longer than that of the WCP-Ph-QSS. Our protocol is highly feasible with current experimental technology and offers a promising approach for long-distance and high-capacity quantum networks.

quant-ph↗

arXiv2026

On consistency around a $3 \times 3\times 3$ cube and Q3 analogue of the lattice Boussinesq equation

In this paper, we present two new aspects of lattice Boussinesq (BSQ) equations. First, we show that the lattice potential BSQ (lpBSQ) equation defined on a nine-point square lattice admits a natural extension of three-dimensional consistency to a $3\times 3\times 3$ cube\textemdash a cubic sublattice consisting of $27$ vertices. This extends the standard notion of three-dimensional consistency (defined on an elementary $2\times 2\times 2$ vertex cube for quadrilateral equations) to the non-quadrilateral, nine-point setting. Second, we construct a new three-component system which is referred to as the {\em lattice BSQ-Q3 system}, serving as the BSQ analogue of the Q3($δ$) equation in the Adler-Bobenko-Suris (ABS) classification. The construction relies on a gauge transformation between Lax pairs of lpBSQ with the parameter $δ$ arising from a $GL_3$ action. In a degeneration form, the system yields a $PGL_3$-invariant integrable lattice equation that generalises the $PGL_2$-invariant Schwarzian BSQ equation.

nlin.SI↗

arXiv2025

HY-Motion 1.0: Scaling Flow Matching Models for Text-To-Motion Generation

We present HY-Motion 1.0, a series of state-of-the-art, large-scale, motion generation models capable of generating 3D human motions from textual descriptions. HY-Motion 1.0 represents the first successful attempt to scale up Diffusion Transformer (DiT)-based flow matching models to the billion-parameter scale within the motion generation domain, delivering instruction-following capabilities that significantly outperform current open-source benchmarks. Uniquely, we introduce a comprehensive, full-stage training paradigm -- including large-scale pretraining on over 3,000 hours of motion data, high-quality fine-tuning on 400 hours of curated data, and reinforcement learning from both human feedback and reward models -- to ensure precise alignment with the text instruction and high motion quality. This framework is supported by our meticulous data processing pipeline, which performs rigorous motion cleaning and captioning. Consequently, our model achieves the most extensive coverage, spanning over 200 motion categories across 6 major classes. We release HY-Motion 1.0 to the open-source community to foster future research and accelerate the transition of 3D human motion generation models towards commercial maturity.

cs.CV↗

arXiv2025

Defect Engineering for Stabilizing Magnetic and Topological Properties in Mn(Bi1-xSbx)2Te4

MnBi2Te4 is a versatile platform for exploring diverse topological quantum states, yet its potential is hampered by intrinsic antisite defects. While Sb substitution has been employed to tune the Fermi level towards the charge neutral point, it exacerbates the formation of Mn-Sb antisite defects. Here, we address this challenge by combining first-principles calculations with strategic synthesis to systematically investigate and control antisite defects in Mn(Bi1-xSbx)2Te4. Our calculations reveal that increasing antisite defect density progressively destroys the field-forced magnetic Weyl state, eventually driving the system into a trivial magnetic insulator. Motivated by these findings, we develop an optimized chemical vapor transport method, yielding high-quality Mn(Bi1-xSbx)2Te4 crystals with significantly reduced antisite defect density. The emergence of strong Shubnikov-de Haas oscillations in the forced ferromagnetic state and a pronounced anomalous Hall effect near charge neutrality, with opposite signs for n- and p-type samples, confirms the type-II Weyl semimetal nature. These findings underscore the critical role of antisite defects in determining the magnetic and topological properties of Mn(Bi1-xSbx)2Te4 and establish defect engineering via optimized synthesis as a crucial strategy for realizing its exotic magnetic topological states.

cond-mat.mtrl-sci↗

arXiv2025

On the formulation of the NQC variable

The Nijhoff-Quispel-Capel (NQC) equation is a general lattice quadrilateral equation presented in terms of a function $S(a,b)$ where $a$ and $b$ serve as extra parameters. It can be viewed as the counterpart of Q3 equation which is the second top equation in the Adler-Bobenko-Suris list. In this paper, we review some known formulations of the NQC variable $S(a,b)$, such as the Cauchy matrix approach, the eigenfunction approach and via a spectral Wronskian. We also present a new perspective to formulate $S(a,b)$ from the eigenfunctions of a Lax pair of the lattice (non-potential) modified Korteweg de Vries equation. A new Dbar problem is introduced and employed in the derivation.

nlin.SI↗

arXiv2025

OpenView: Empowering MLLMs with Out-of-view VQA

Recent multimodal large language models (MLLMs) show great potential in natural image understanding. Yet, they perform well, mainly on reasoning in-view contents within the image frame. This paper presents the first study on out-of-view (OOV) understanding, i.e., the ability to reason objects, activities, and scenes beyond the visible frame of a perspective view. Our technical contributions are threefold. First, we design OpenView, a four-stage pipeline to massively generate multi-choice VQA by leveraging panoramic imagery to enable context-rich and spatial-grounded VQA synthesis with free-view framing. Second, we curate OpenView-Dataset, a high-quality synthetic dataset from diverse real-world panoramas to empower MLLMs upon supervised fine-tuning. Third, we build OpenView-Bench, a benchmark that jointly measures choice and rationale accuracy for interpretable and diagnosable evaluation. Experimental results show that despite having a large gap from human performance in OOV VQA answer selection, upon empowered by OpenView, multiple MLLMs can consistently boost their performance, uplifted from 48.6% to 64.1% on average. Code, benchmark, and data will be available at https://github.com/q1xiangchen/OpenView.

cs.CV↗

arXiv2025

ASAP-Textured Gaussians: Enhancing Textured Gaussians with Adaptive Sampling and Anisotropic Parameterization

Recent advances have equipped 3D Gaussian Splatting with texture parameterizations to capture spatially varying attributes, improving the performance of both appearance modeling and downstream tasks. However, the added texture parameters introduce significant memory efficiency challenges. Rather than proposing new texture formulations, we take a step back to examine the characteristics of existing textured Gaussian methods and identify two key limitations in common: (1) Textures are typically defined in canonical space, leading to inefficient sampling that wastes textures' capacity on low-contribution regions; and (2) texture parameterization is uniformly assigned across all Gaussians, regardless of their visual complexity, resulting in over-parameterization. In this work, we address these issues through two simple yet effective strategies: adaptive sampling based on the Gaussian density distribution and error-driven anisotropic parameterization that allocates texture resources according to rendering error. Our proposed ASAP Textured Gaussians, short for Adaptive Sampling and Anisotropic Parameterization, significantly improve the quality efficiency tradeoff, achieving high-fidelity rendering with far fewer texture parameters.

cs.CV↗

arXiv2025

Beyond Training: Enabling Self-Evolution of Agents with MOBIMEM

Large Language Model (LLM) agents are increasingly deployed to automate complex workflows in mobile and desktop environments. However, current model-centric agent architectures struggle to self-evolve post-deployment: improving personalization, capability, and efficiency typically requires continuous model retraining/fine-tuning, which incurs prohibitive computational overheads and suffers from an inherent trade-off between model accuracy and inference efficiency. To enable iterative self-evolution without model retraining, we propose MOBIMEM, a memory-centric agent system. MOBIMEM first introduces three specialized memory primitives to decouple agent evolution from model weights: (1) Profile Memory uses a lightweight distance-graph (DisGraph) structure to align with user preferences, resolving the accuracy-latency trade-off in user profile retrieval; (2) Experience Memory employs multi-level templates to instantiate execution logic for new tasks, ensuring capability generalization; and (3) Action Memory records fine-grained interaction sequences, reducing the reliance on expensive model inference. Building upon this memory architecture, MOBIMEM further integrates a suite of OS-inspired services to orchestrate execution: a scheduler that coordinates parallel sub-task execution and memory operations; an agent record-and-replay (AgentRR) mechanism that enables safe and efficient action reuse; and a context-aware exception handling that ensures graceful recovery from user interruptions and runtime errors. Evaluation on AndroidWorld and top-50 apps shows that MOBIMEM achieves 83.1% profile alignment with 23.83 ms retrieval time (280x faster than GraphRAG baselines), improves task success rates by up to 50.3%, and reduces end-to-end latency by up to 9x on mobile devices.

cs.AI↗

arXiv2025

Purified pseudomode model for nonlinear system-bath interactions

The theory of purified pseudomodes [arXiv:2412.04264 (2024)] was recently developed to provide a numerical tool for the analysis of the properties of a quantum system and the environment it couples to via linear system-bath interactions. Here we extend this theory to allow for the description of general nonlinear system-bath interactions. We demonstrate the validity of our method by considering the spontaneous decay of a two-level atom placed inside a single-mode lossy cavity and furthermore, its potential application to nanophotonics by calculating the resonance fluorescence spectrum of a quantum dot in the presence of a phonon environment. Our method provides a useful tool for the study of phonon-assisted emission in quantum dots and holds the the promise for broad applications in fields like quantum biology, nonlinear phononics, and nanophotonics.

quant-ph↗

arXiv2025

PanFlow: Decoupled Motion Control for Panoramic Video Generation

Panoramic video generation has attracted growing attention due to its applications in virtual reality and immersive media. However, existing methods lack explicit motion control and struggle to generate scenes with large and complex motions. We propose PanFlow, a novel approach that exploits the spherical nature of panoramas to decouple the highly dynamic camera rotation from the input optical flow condition, enabling more precise control over large and dynamic motions. We further introduce a spherical noise warping strategy to promote loop consistency in motion across panorama boundaries. To support effective training, we curate a large-scale, motion-rich panoramic video dataset with frame-level pose and flow annotations. We also showcase the effectiveness of our method in various applications, including motion transfer and video editing. Extensive experiments demonstrate that PanFlow significantly outperforms prior methods in motion fidelity, visual quality, and temporal coherence. Our code, dataset, and models are available at https://github.com/chengzhag/PanFlow.

cs.CV↗

arXiv2025

"Why the face?": Exploring Robot Error Detection Using Instrumented Bystander Reactions

How do humans recognize and rectify social missteps? We achieve social competence by looking around at our peers, decoding subtle cues from bystanders - a raised eyebrow, a laugh - to evaluate the environment and our actions. Robots, however, struggle to perceive and make use of these nuanced reactions. By employing a novel neck-mounted device that records facial expressions from the chin region, we explore the potential of previously untapped data to capture and interpret human responses to robot error. First, we develop NeckNet-18, a 3D facial reconstruction model to map the reactions captured through the chin camera onto facial points and head motion. We then use these facial responses to develop a robot error detection model which outperforms standard methodologies such as using OpenFace or video data, generalizing well especially for within-participant data. Through this work, we argue for expanding human-in-the-loop robot sensing, fostering more seamless integration of robots into diverse human environments, pushing the boundaries of social cue detection and opening new avenues for adaptable robotics.

cs.RO↗

arXiv2025

Cook and Clean Together: Teaching Embodied Agents for Parallel Task Execution

Task scheduling is critical for embodied AI, enabling agents to follow natural language instructions and execute actions efficiently in 3D physical worlds. However, existing datasets often simplify task planning by ignoring operations research (OR) knowledge and 3D spatial grounding. In this work, we propose Operations Research knowledge-based 3D Grounded Task Scheduling (ORS3D), a new task that requires the synergy of language understanding, 3D grounding, and efficiency optimization. Unlike prior settings, ORS3D demands that agents minimize total completion time by leveraging parallelizable subtasks, e.g., cleaning the sink while the microwave operates. To facilitate research on ORS3D, we construct ORS3D-60K, a large-scale dataset comprising 60K composite tasks across 4K real-world scenes. Furthermore, we propose GRANT, an embodied multi-modal large language model equipped with a simple yet effective scheduling token mechanism to generate efficient task schedules and grounded actions. Extensive experiments on ORS3D-60K validate the effectiveness of GRANT across language understanding, 3D grounding, and scheduling efficiency. The code is available at https://github.com/H-EmbodVis/GRANT

cs.CV↗

arXiv2025

Geometry-induced azimuthal anisotropy in coherent $J/ψ$ photoproduction

Azimuthal anisotropies in heavy-ion collisions are conventionally interpreted as signatures of hydrodynamic flow. We demonstrate that in peripheral collisions, a significant $\cos 2ϕ$ asymmetry in the decay leptons of coherently photoproduced $J/ψ$ mesons arises purely from the initial-state geometry of the nuclear electromagnetic field. This modulation originates from the linear polarization of coherent photons, which is radially aligned in impact parameter space and transferred to the vector meson. By employing light-cone perturbation theory within the dipole formalism, we calculate the centrality dependence of this asymmetry for collisions at RHIC and LHC energies. Our predictions quantitatively reproduce STAR data. This observable thus provides a rigorous benchmark for distinguishing electromagnetic initial-state effects from collective medium dynamics.

hep-ph↗

arXiv2025

SEMC: Structure-Enhanced Mixture-of-Experts Contrastive Learning for Ultrasound Standard Plane Recognition

Ultrasound standard plane recognition is essential for clinical tasks such as disease screening, organ evaluation, and biometric measurement. However, existing methods fail to effectively exploit shallow structural information and struggle to capture fine-grained semantic differences through contrastive samples generated by image augmentations, ultimately resulting in suboptimal recognition of both structural and discriminative details in ultrasound standard planes. To address these issues, we propose SEMC, a novel Structure-Enhanced Mixture-of-Experts Contrastive learning framework that combines structure-aware feature fusion with expert-guided contrastive learning. Specifically, we first introduce a novel Semantic-Structure Fusion Module (SSFM) to exploit multi-scale structural information and enhance the model's ability to perceive fine-grained structural details by effectively aligning shallow and deep features. Then, a novel Mixture-of-Experts Contrastive Recognition Module (MCRM) is designed to perform hierarchical contrastive learning and classification across multi-level features using a mixture-of-experts (MoE) mechanism, further improving class separability and recognition performance. More importantly, we also curate a large-scale and meticulously annotated liver ultrasound dataset containing six standard planes. Extensive experimental results on our in-house dataset and two public datasets demonstrate that SEMC outperforms recent state-of-the-art methods across various metrics.

cs.CV↗

arXiv2025

Provable Sample-Efficient Transfer Learning Conditional Diffusion Models via Representation Learning

While conditional diffusion models have achieved remarkable success in various applications, they require abundant data to train from scratch, which is often infeasible in practice. To address this issue, transfer learning has emerged as an essential paradigm in small data regimes. Despite its empirical success, the theoretical underpinnings of transfer learning conditional diffusion models remain unexplored. In this paper, we take the first step towards understanding the sample efficiency of transfer learning conditional diffusion models through the lens of representation learning. Inspired by practical training procedures, we assume that there exists a low-dimensional representation of conditions shared across all tasks. Our analysis shows that with a well-learned representation from source tasks, the samplecomplexity of target tasks can be reduced substantially. In addition, we investigate the practical implications of our theoretical results in several real-world applications of conditional diffusion models. Numerical experiments are also conducted to verify our results.

cs.LG↗

arXiv2025

A Framework for Hybrid Physics-AI Coupled Ocean Models

Climate simulations, at all grid resolutions, rely on approximations that encapsulate the forcing due to unresolved processes on resolved variables, known as parameterizations. Parameterizations often lead to inaccuracies in climate models, with significant biases in the physics of key climate phenomena. Advances in artificial intelligence (AI) are now directly enabling the learning of unresolved processes from data to improve the physics of climate simulations. Here, we introduce a flexible framework for developing and implementing physics- and scale-aware machine learning parameterizations within climate models. We focus on the ocean and sea-ice components of a state-of-the-art climate model by implementing a spectrum of data-driven parameterizations, ranging from complex deep learning models to more interpretable equation-based models. Our results showcase the viability of AI-driven parameterizations in operational models, advancing the capabilities of a new generation of hybrid simulations, and include prototypes of fully coupled atmosphere-ocean-sea-ice hybrid simulations. The tools developed are open source, accessible, and available to all.

physics.ao-ph↗