Searcharxiv⌕ Search

arXiv subjects

Bo Zhu

Publications and source records attributed to Bo Zhu.

At least 55 records · Page 3Linked to original sources
arXiv2025

Skywork UniPic: Unified Autoregressive Modeling for Visual Understanding and Generation

We introduce Skywork UniPic, a 1.5 billion-parameter autoregressive model that unifies image understanding, text-to-image generation, and image editing within a single architecture-eliminating the need for task-specific adapters or inter-module connectors-and demonstrate that compact multimodal systems can achieve state-of-the-art performance on commodity hardware. Skywork UniPic achieves a GenEval score of 0.86, surpassing most existing unified models; sets a new DPG-Bench complex-generation record of 85.5; attains 5.83 on GEditBench-EN and 3.49 on ImgEdit-Bench for image editing; and generates 1024 x 1024 images with under 15 GB of GPU memory (e.g., RTX 4090). (1) a decoupled encoding strategy that leverages a masked autoregressive encoder for synthesis and a SigLIP2 encoder for understanding, all feeding a shared autoregressive decoder; (2) a progressive, resolution-aware training schedule scaling from 256 x 256 to 1024 x 1024 while dynamically unfreezing parameters to balance capacity and stability; and (3) meticulously curated, 100 million-scale datasets augmented with task-specific reward models to refine generation and editing objectives. By demonstrating that high-fidelity multimodal integration need not incur prohibitive resource demands, Skywork UniPic establishes a practical paradigm for deployable, high-fidelity multimodal AI. Code and weights are publicly available at https://huggingface.co/Skywork/Skywork-UniPic-1.5B.

cs.CV↗
arXiv2025

FluidNexus: 3D Fluid Reconstruction and Prediction from a Single Video

We study reconstructing and predicting 3D fluid appearance and velocity from a single video. Current methods require multi-view videos for fluid reconstruction. We present FluidNexus, a novel framework that bridges video generation and physics simulation to tackle this task. Our key insight is to synthesize multiple novel-view videos as references for reconstruction. FluidNexus consists of two key components: (1) a novel-view video synthesizer that combines frame-wise view synthesis with video diffusion refinement for generating realistic videos, and (2) a physics-integrated particle representation coupling differentiable simulation and rendering to simultaneously facilitate 3D fluid reconstruction and prediction. To evaluate our approach, we collect two new real-world fluid datasets featuring textured backgrounds and object interactions. Our method enables dynamic novel view synthesis, future prediction, and interaction simulation from a single fluid video. Project website: https://yuegao.me/FluidNexus.

cs.CV↗
arXiv2025

Networked pointing system: Bearing-only target localization and pointing control

In the paper, we formulate the target-pointing consensus problem where the headings of agents are required to point at a common target. Only a few agents in the network can measure the bearing information of the target. A two-step solution consisting of a bearing-only estimator for target localization and a control law for target pointing is constructed to address this problem. Compared to the strong assumptions of existing works, we only require two agents not collinear with the target to ensure localizability. By introducing the concept of virtual fusion node, we prove that both the estimation error and the tracking error converge asymptotically to the origin. The video demonstration of the verification can be found at https://youtu.be/S9- eyofk1DY.

eess.SY↗
arXiv2025

Fluid Simulation on Vortex Particle Flow Maps

We propose the Vortex Particle Flow Map (VPFM) method to simulate incompressible flow with complex vortical evolution in the presence of dynamic solid boundaries. The core insight of our approach is that vorticity is an ideal quantity for evolution on particle flow maps, enabling significantly longer flow map distances compared to other fluid quantities like velocity or impulse. To achieve this goal, we developed a hybrid Eulerian-Lagrangian representation that evolves vorticity and flow map quantities on vortex particles, while reconstructing velocity on a background grid. The method integrates three key components: (1) a vorticity-based particle flow map framework, (2) an accurate Hessian evolution scheme on particles, and (3) a solid boundary treatment for no-through and no-slip conditions in VPFM. These components collectively allow a substantially longer flow map length (3-12 times longer) than the state-of-the-art, enhancing vorticity preservation over extended spatiotemporal domains. We validated the performance of VPFM through diverse simulations, demonstrating its effectiveness in capturing complex vortex dynamics and turbulence phenomena.

cs.GR↗
arXiv2025

PID-GM: PID Control with Gain Mapping

Proportional-Integral-Differential (PID) control is widely used in industrial control systems. However, up to now there are at least two open problems related with PID control. One is to have a comprehensive understanding of its robustness with respect to model uncertainties and disturbances. The other is to build intuitive, explicit and mathematically provable guidelines for PID gain tuning. In this paper, we introduce a simple nonlinear mapping to determine PID gains from three auxiliary parameters. By the mapping, PID control is shown to be equivalent to a new PD control (serving as a nominal control) plus an uncertainty and disturbance compensator (to recover the nominal performance). Then PID control can be understood, designed and tuned in a Two-Degree-of-Freedom (2-DoF) control framework. We discuss some basic properties of the mapping, including the existence, uniqueness and invertibility. Taking as an example the PID control applied to a general uncertain second-order plant, we prove by the singular perturbation theory that the closed-loop steady-state and transient performance depends explicitly on one auxiliary parameter which can be viewed as the virtual singular perturbation parameter (SPP) of PID control. All the three PID gains are monotonically decreasing functions of the SPP, indicating that the smaller the SPP is, the higher the PID gains are, and the better the robustness of PID control is. Simulation and experimental examples are provided to demonstrate the properties of the mapping as well as the effectiveness of the mapping based PID gain turning.

eess.SY↗
arXiv2025

Scalar-mean rigidity theorem for compact manifolds with boundary

We prove a scalar-mean rigidity theorem for compact Riemannian manifolds with boundary in dimension less than five by developing a dimension reduction argument for mean curvature, which extends Schoen-Yau's dimension reduction argument for scalar curvature. As a corollary, we prove the sharp spherical radius rigidity theorem and best NNSC fill-in in terms of the mean curvature. Moreover, we prove a Lipschitz Listing type scalar-mean rigidity theorem for these dimensions.

math.DG↗
arXiv2024

On the Exploration of LM-Based Soft Modular Robot Design

Recent large language models (LLMs) have demonstrated promising capabilities in modeling real-world knowledge and enhancing knowledge-based generation tasks. In this paper, we further explore the potential of using LLMs to aid in the design of soft modular robots, taking into account both user instructions and physical laws, to reduce the reliance on extensive trial-and-error experiments typically needed to achieve robot designs that meet specific structural or task requirements. Specifically, we formulate the robot design process as a sequence generation task and find that LLMs are able to capture key requirements expressed in natural language and reflect them in the construction sequences of robots. To simplify, rather than conducting real-world experiments to assess design quality, we utilize a simulation tool to provide feedback to the generative model, allowing for iterative improvements without requiring extensive human annotations. Furthermore, we introduce five evaluation metrics to assess the quality of robot designs from multiple angles including task completion and adherence to instructions, supporting an automatic evaluation process. Our model performs well in evaluations for designing soft modular robots with uni- and bi-directional locomotion and stair-descending capabilities, highlighting the potential of using natural language and LLMs for robot design. However, we also observe certain limitations that suggest areas for further improvement.

cs.RO↗
arXiv2024

NeuralFluid: Neural Fluidic System Design and Control with Differentiable Simulation

We present a novel framework to explore neural control and design of complex fluidic systems with dynamic solid boundaries. Our system features a fast differentiable Navier-Stokes solver with solid-fluid interface handling, a low-dimensional differentiable parametric geometry representation, a control-shape co-design algorithm, and gym-like simulation environments to facilitate various fluidic control design applications. Additionally, we present a benchmark of design, control, and learning tasks on high-fidelity, high-resolution dynamic fluid environments that pose challenges for existing differentiable fluid simulators. These tasks include designing the control of artificial hearts, identifying robotic end-effector shapes, and controlling a fluid gate. By seamlessly incorporating our differentiable fluid simulator into a learning framework, we demonstrate successful design, control, and learning results that surpass gradient-free solutions in these benchmark tasks.

physics.flu-dyn↗
arXiv2024

Positive Scalar Curvature Meets Ricci Limit Spaces

We investigate the influence of uniformly positive scalar curvature on the size of a non-collapsed Ricci limit space coming from a sequence of $n$-manifolds with non-negative Ricci curvature and uniformly positive scalar curvature. We prove that such a limit space splits at most $n-2$ lines or $\mathbb{R}$-factors. When this maximal splitting occurs, we obtain a uniform upper bound on the diameter of the non-splitting factor. Moreover, we obtain a volume gap estimate and a volume growth order estimate of geodesic balls on such manifolds.

math.DG↗
arXiv2024

An Interface Tracking Method with Triangle Edge Cuts

This paper introduces a volume-conserving interface tracking algorithm on unstructured triangle meshes. We propose to discretize the interface via triangle edge cuts which represent the intersections between the interface and the triangle mesh edges using a compact 6 numbers per triangle. This enables an efficient implicit representation of the sub-triangle polygonal material regions without explicitly storing connectivity information. Moreover, we propose an efficient advection algorithm for this interface representation that is based on geometric queries and does not require an optimization process. This advection algorithm is extended via an area correction step that enforces volume-conservation of the materials. We demonstrate the efficacy of our method on a variety of advection problems on a triangle mesh and compare its performance to existing interface tracking methods including VOF and MOF.

cs.GR↗
arXiv2024

Bounding the A-hat genus using scalar curvature lower bounds and isoperimetric constants

In this paper, we prove an upper bound on the $\widehat{A}$ genus of a smooth, closed, spin Riemannian manifold using its scalar curvature lower bound, Neumann isoperimetric constant, and volume. The proof of this result relies on spectral analysis of the Dirac operator. We also construct an example to show that the Neumann isoperimetric constant in this bound is necessary. Our result partially answers a question of Gromov on bounding characteristic numbers using scalar curvature lower bound.

math.DG↗
arXiv2024

MoE++: Accelerating Mixture-of-Experts Methods with Zero-Computation Experts

In this work, we aim to simultaneously enhance the effectiveness and efficiency of Mixture-of-Experts (MoE) methods. To achieve this, we propose MoE++, a general and heterogeneous MoE framework that integrates both Feed-Forward Network~(FFN) and zero-computation experts. Specifically, we introduce three types of zero-computation experts: the zero expert, copy expert, and constant expert, which correspond to discard, skip, and replace operations, respectively. This design offers three key advantages: (i) Low Computing Overhead: Unlike the uniform mixing mechanism for all tokens within vanilla MoE, MoE++ allows each token to engage with a dynamic number of FFNs, be adjusted by constant vectors, or even skip the MoE layer entirely. (ii) High Performance: By enabling simple tokens to utilize fewer FFN experts, MoE++ allows more experts to focus on challenging tokens, thereby unlocking greater performance potential than vanilla MoE. (iii) Deployment Friendly: Given that zero-computation experts have negligible parameters, we can deploy all zero-computation experts on each GPU, eliminating the significant communication overhead and expert load imbalance associated with FFN experts distributed across different GPUs. Moreover, we leverage gating residuals, enabling each token to consider the pathway taken in the previous layer when selecting the appropriate experts. Extensive experimental results demonstrate that MoE++ achieves better performance while delivering 1.1-2.1x expert forward throughput compared to a vanilla MoE model of the same size, which lays a solid foundation for developing advanced and efficient MoE-related models.

cs.LG↗
arXiv2024

An Eulerian Vortex Method on Flow Maps

We present an Eulerian vortex method based on the theory of flow maps to simulate the complex vortical motions of incompressible fluids. Central to our method is the novel incorporation of the flow-map transport equations for line elements, which, in combination with a bi-directional marching scheme for flow maps, enables the high-fidelity Eulerian advection of vorticity variables. The fundamental motivation is that, compared to impulse $\mathbf{m}$, which has been recently bridged with flow maps to encouraging results, vorticity $\boldsymbolω$ promises to be preferable for its numerical stability and physical interpretability. To realize the full potential of this novel formulation, we develop a new Poisson solving scheme for vorticity-to-velocity reconstruction that is both efficient and able to accurately handle the coupling near solid boundaries. We demonstrate the efficacy of our approach with a range of vortex simulation examples, including leapfrog vortices, vortex collisions, cavity flow, and the formation of complex vortical structures due to solid-fluid interactions.

cs.GR↗
arXiv2024

Solid-Fluid Interaction on Particle Flow Maps

We propose a novel solid-fluid interaction method for coupling elastic solids with impulse flow maps. Our key idea is to unify the representation of fluid and solid components as particle flow maps with different lengths and dynamics. The solid-fluid coupling is enabled by implementing two novel mechanisms: first, we developed an impulse-to-velocity transfer mechanism to unify the exchanged physical quantities; second, we devised a particle path integral mechanism to accumulate coupling forces along each flow-map trajectory. Our framework integrates these two mechanisms into an Eulerian-Lagrangian impulse fluid simulator to accommodate traditional coupling models, exemplified by the Material Point Method (MPM) and Immersed Boundary Method (IBM), within a particle flow map framework. We demonstrate our method's efficacy by simulating solid-fluid interactions exhibiting strong vortical dynamics, including various vortex shedding and interaction examples across swimming, falling, breezing, and combustion.

cs.GR↗
arXiv2024

Particle-Laden Fluid on Flow Maps

We propose a novel framework for simulating ink as a particle-laden flow using particle flow maps. Our method addresses the limitations of existing flow-map techniques, which struggle with dissipative forces like viscosity and drag, thereby extending the application scope from solving the Euler equations to solving the Navier-Stokes equations with accurate viscosity and laden-particle treatment. Our key contribution lies in a coupling mechanism for two particle systems, coupling physical sediment particles and virtual flow-map particles on a background grid by solving a Poisson system. We implemented a novel path integral formula to incorporate viscosity and drag forces into the particle flow map process. Our approach enables state-of-the-art simulation of various particle-laden flow phenomena, exemplified by the bulging and breakup of suspension drop tails, torus formation, torus disintegration, and the coalescence of sedimenting drops. In particular, our method delivered high-fidelity ink diffusion simulations by accurately capturing vortex bulbs, viscous tails, fractal branching, and hierarchical structures.

cs.GR↗
arXiv2024

Heuristic Predictive Control for Multi-Robot Flocking in Congested Environments

Multi-robot flocking possesses extraordinary advantages over a single-robot system in diverse domains, but it is challenging to ensure safe and optimal performance in congested environments. Hence, this paper is focused on the investigation of distributed optimal flocking control for multiple robots in crowded environments. A heuristic predictive control solution is proposed based on a Gibbs Random Field (GRF), in which bio-inspired potential functions are used to characterize robot-robot and robot-environment interactions. The optimal solution is obtained by maximizing a posteriori joint distribution of the GRF in a certain future time instant. A gradient-based heuristic solution is developed, which could significantly speed up the computation of the optimal control. Mathematical analysis is also conducted to show the validity of the heuristic solution. Multiple collision risk levels are designed to improve the collision avoidance performance of robots in dynamic environments. The proposed heuristic predictive control is evaluated comprehensively from multiple perspectives based on different metrics in a challenging simulation environment. The competence of the proposed algorithm is validated via the comparison with the non-heuristic predictive control and two existing popular flocking control methods. Real-life experiments are also performed using four quadrotor UAVs to further demonstrate the efficiency of the proposed design.

cs.RO↗
arXiv2024

Skywork-MoE: A Deep Dive into Training Techniques for Mixture-of-Experts Language Models

In this technical report, we introduce the training methodologies implemented in the development of Skywork-MoE, a high-performance mixture-of-experts (MoE) large language model (LLM) with 146 billion parameters and 16 experts. It is initialized from the pre-existing dense checkpoints of our Skywork-13B model. We explore the comparative effectiveness of upcycling versus training from scratch initializations. Our findings suggest that the choice between these two approaches should consider both the performance of the existing dense checkpoints and the MoE training budget. We highlight two innovative techniques: gating logit normalization, which improves expert diversification, and adaptive auxiliary loss coefficients, allowing for layer-specific adjustment of auxiliary loss coefficients. Our experimental results validate the effectiveness of these methods. Leveraging these techniques and insights, we trained our upcycled Skywork-MoE on a condensed subset of our SkyPile corpus. The evaluation results demonstrate that our model delivers strong performance across a wide range of benchmarks.

cs.CL↗