arXiv subjects

Yi Peng

Publications and source records attributed to Yi Peng.

At least 37 records · Page 2Linked to original sources

arXiv2025

Confinement reduces surface accumulation of swimming bacteria

Many swimming bacteria naturally inhabit confined environments, yet how confinement influences their swimming behaviors remains unclear. Here, we combine experiments, continuum modeling and particle-based simulations to investigate near-surface bacterial swimming in dilute suspensions under varying confinement. Confinement reduces near-surface accumulation and facilitates bacterial escape. These effects are quantitatively captured by models incorporating the force quadrupole, a higher-order hydrodynamic singularity, that generates a rotational flow reorienting bacteria away from surfaces. Under strong confinement, bacterial trajectories straighten due to the balancing torques exerted by opposing surfaces. These findings highlight the role of hydrodynamic quadrupole interactions in near-surface bacterial motility, with implications for microbial ecology, infection control, and industrial applications.

physics.bio-ph↗

arXiv2025

Non-uniqueness of weak solutions to the 3D Hall-MHD equations on the plane

We prove the non-uniqueness of weak solutions with non-trivial magnetic fields to the 3D Hall-MHD equations on the plane in the space $C^0_t L_x^2$ through the convex integration scheme and by constructing new errors and new intermittent flows. In particular, based on the construction of 3D intermittent flows, we obtain the $2\frac{1}{2}$D Mikado flows through a projection onto the plane. Moreover, we prove that the constructed weak solution do not conserve the magnetic helicity and find that weak solutions of the ideal Hall-MHD equations in $C^{\barβ}_{t,x}$ ($\barβ>0$) are the strong vanishing viscosity and resistive limit of weak solutions to the Hall-MHD equations.

math.AP↗

arXiv2025

CKFNet: Neural Network Aided Cubature Kalman filtering

The cubature Kalman filter (CKF), while theoretically rigorous for nonlinear estimation, often suffers performance degradation due to model-environment mismatches in practice. To address this limitation, we propose CKFNet-a hybrid architecture that synergistically integrates recurrent neural networks (RNN) with the CKF framework while preserving its cubature principles. Unlike conventional model-driven approaches, CKFNet embeds RNN modules in the prediction phase to dynamically adapt to unmodeled uncertainties, effectively reducing cumulative error propagation through temporal noise correlation learning. Crucially, the architecture maintains CKF's analytical interpretability via constrained optimization of cubature point distributions. Numerical simulation experiments have confirmed that our proposed CKFNet exhibits superior accuracy and robustness compared to conventional model-based methods and existing KalmanNet algorithms.

eess.SP↗

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

Skywork-R1V3 Technical Report

We introduce Skywork-R1V3, an advanced, open-source vision-language model (VLM) that pioneers a new approach to visual reasoning. Its key innovation lies in effectively transferring reasoning skills from text-only Large Language Models (LLMs) to visual tasks. The strong performance of Skywork-R1V3 primarily stems from our elaborate post-training RL framework, which effectively activates and enhances the model's reasoning ability, without the need for additional continue pre-training. Through this framework, we further uncover the fundamental role of the connector module in achieving robust cross-modal alignment for multimodal reasoning models. In addition, we introduce a unique indicator of reasoning capability, the entropy of critical reasoning tokens, which has proven highly effective for checkpoint selection during RL training. Skywork-R1V3 achieves state-of-the-art results on MMMU, significantly improving from 64.3% to 76.0%. This performance matches entry-level human capabilities. Remarkably, our RL-powered post-training approach enables even the 38B parameter model to rival top closed-source VLMs. The implementation successfully transfers mathematical reasoning to other subject-related reasoning tasks. We also include an analysis of curriculum learning and reinforcement finetuning strategies, along with a broader discussion on multimodal reasoning. Skywork-R1V3 represents a significant leap in multimodal reasoning, showcasing RL as a powerful engine for advancing open-source VLM capabilities.

cs.CL↗

arXiv2025

Phase Transition in Non-isentropic Compressible Immiscible Two-Phase Flow with van der Waals Equation of State

This study establishes the global well-posedness of the compressible non-isentropic Navier-Stokes/Allen-Cahn system governed by the van der Waals equation of state $p(ρ,θ)=- aρ^2+\frac{Rθρ}{1-bρ}$ and degenerate thermal conductivity $κ(θ)=\tildeκθ^β$, where $p$, $ρ$ and $θ$ are the pressure, the density and the temperature of the flow respectively, and $a,b,R,\tildeκ$ are positive constants related to the physical properties of the flow. Navier-Stokes/Allen-Cahn system models immiscible two-phase flow with diffusive interfaces, where the non-monotonic pressure-density relationship in the van der Waals equation drives gas-liquid phase transitions. By developing a refined $L^2$-energy framework, we prove the existence and uniqueness of global strong solutions to the one-dimensional Cauchy problem for non-vacuum and finite-temperature initial data, without imposing smallness restrictions on the initial conditions. The findings demonstrate that despite non-monotonic pressure inducing substantial density fluctuations and triggering phase transitions, all physical quantities remain bounded over finite time intervals.

math.AP↗

arXiv2025

CSVQA: A Chinese Multimodal Benchmark for Evaluating STEM Reasoning Capabilities of VLMs

Vision-Language Models (VLMs) have demonstrated remarkable progress in multimodal understanding, yet their capabilities for scientific reasoning remain inadequately assessed. Current multimodal benchmarks predominantly evaluate generic image comprehension or text-driven reasoning, lacking authentic scientific contexts that require domain-specific knowledge integration with visual evidence analysis. To fill this gap, we present CSVQA, a diagnostic multimodal benchmark specifically designed for evaluating scientific reasoning through domain-grounded visual question answering. Our benchmark features 1,378 carefully constructed question-answer pairs spanning diverse STEM disciplines, each demanding domain knowledge, integration of visual evidence, and higher-order reasoning. Compared to prior multimodal benchmarks, CSVQA places greater emphasis on real-world scientific content and complex reasoning. We additionally propose a rigorous evaluation protocol to systematically assess whether model predictions are substantiated by valid intermediate reasoning steps based on curated explanations. Our comprehensive evaluation of 15 VLMs on this benchmark reveals notable performance disparities, as even the top-ranked proprietary model attains only 49.6% accuracy. This empirical evidence underscores the pressing need for advancing scientific reasoning capabilities in VLMs. Our CSVQA is released at https://huggingface.co/datasets/Skywork/CSVQA

cs.CV↗

arXiv2025

Skywork R1V: Pioneering Multimodal Reasoning with Chain-of-Thought

We introduce Skywork R1V, a multimodal reasoning model extending the an R1-series Large language models (LLM) to visual modalities via an efficient multimodal transfer method. Leveraging a lightweight visual projector, Skywork R1V facilitates seamless multimodal adaptation without necessitating retraining of either the foundational language model or the vision encoder. To strengthen visual-text alignment, we propose a hybrid optimization strategy that combines Iterative Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO), significantly enhancing cross-modal integration efficiency. Additionally, we introduce an adaptive-length Chain-of-Thought distillation approach for reasoning data generation. This approach dynamically optimizes reasoning chain lengths, thereby enhancing inference efficiency and preventing excessive reasoning overthinking. Empirical evaluations demonstrate that Skywork R1V, with only 38B parameters, delivers competitive performance, achieving a score of 69.0 on the MMMU benchmark and 67.5 on MathVista. Meanwhile, it maintains robust textual reasoning performance, evidenced by impressive scores of 72.0 on AIME and 94.0 on MATH500. The Skywork R1V model weights have been publicly released to promote openness and reproducibility.

cs.CV↗

arXiv2025

Skywork-VL Reward: An Effective Reward Model for Multimodal Understanding and Reasoning

We propose Skywork-VL Reward, a multimodal reward model that provides reward signals for both multimodal understanding and reasoning tasks. Our technical approach comprises two key components: First, we construct a large-scale multimodal preference dataset that covers a wide range of tasks and scenarios, with responses collected from both standard vision-language models (VLMs) and advanced VLM reasoners. Second, we design a reward model architecture based on Qwen2.5-VL-7B-Instruct, integrating a reward head and applying multi-stage fine-tuning using pairwise ranking loss on pairwise preference data. Experimental evaluations show that Skywork-VL Reward achieves state-of-the-art results on multimodal VL-RewardBench and exhibits competitive performance on the text-only RewardBench benchmark. Furthermore, preference data constructed based on our Skywork-VL Reward proves highly effective for training Mixed Preference Optimization (MPO), leading to significant improvements in multimodal reasoning capabilities. Our results underscore Skywork-VL Reward as a significant advancement toward general-purpose, reliable reward models for multimodal alignment. Our model has been publicly released to promote transparency and reproducibility.

cs.CV↗

arXiv2025

Skywork R1V2: Multimodal Hybrid Reinforcement Learning for Reasoning

We present Skywork R1V2, a next-generation multimodal reasoning model and a major leap forward from its predecessor, Skywork R1V. At its core, R1V2 introduces a hybrid reinforcement learning paradigm that jointly leverages the Mixed Preference Optimization (MPO) and the Group Relative Policy Optimization (GRPO), which harmonizes reward-model guidance with rule-based strategies, thereby addressing the long-standing challenge of balancing sophisticated reasoning capabilities with broad generalization. To further enhance training efficiency, we propose the Selective Sample Buffer (SSB) mechanism, which effectively addresses the vanishing advantages dilemma inherent in GRPO by prioritizing high-value samples throughout the optimization process. Notably, we observe that excessive reinforcement signals can induce visual hallucinations--a phenomenon we systematically monitor and mitigate through calibrated reward thresholds throughout the training process. Empirical results affirm the exceptional capability of R1V2, with benchmark-leading performances such as 62.6 on OlympiadBench, 78.9 on AIME2024, 63.6 on LiveCodeBench, and 73.6 on MMMU. These results underscore R1V2's superiority over existing open-source models and demonstrate significant progress in closing the performance gap with premier proprietary systems, including Gemini 2.5 and OpenAI-o4-mini. The Skywork R1V2 model weights have been publicly released to promote openness and reproducibility https://huggingface.co/Skywork/Skywork-R1V2-38B.

cs.CV↗

arXiv2025

A Family of Robust Generalized Adaptive Filters and Application for Time-series Prediction

The continuous development of new adaptive filters (AFs) based on novel cost functions (CFs) is driven by the demands of various application scenarios and noise environments. However, these algorithms typically demonstrate optimal performance only in specific conditions. In the event of the noise change, the performance of these AFs often declines, rendering simple parameter adjustments ineffective. Instead, a modification of the CF is necessary. To address this issue, the robust generalized adaptive AF (RGA-AF) with strong adaptability and flexibility is proposed in this paper. The flexibility of the RGA-AF's CF allows for smooth adaptation to varying noise environments through parameter adjustments, ensuring optimal filtering performance in diverse scenarios. Moreover, we introduce several fundamental properties of negative RGA (NRGA) entropy and present the negative asymmetric RGA-AF (NAR-GA-AF) and kernel recursive NRGA-AF (KRNRGA-AF). These AFs address asymmetric noise distribution and nonlinear filtering issues, respectively. Simulations of linear system identification and time-series prediction for Chua's circuit under different noise environments demonstrate the superiority of the proposed algorithms in comparison to existing techniques.

eess.SP↗

arXiv2025

HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformer

Recent advancements in image generative foundation models have prioritized quality improvements but often at the cost of increased computational complexity and inference latency. To address this critical trade-off, we introduce HiDream-I1, a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within seconds. HiDream-I1 is constructed with a new sparse Diffusion Transformer (DiT) structure. Specifically, it starts with a dual-stream decoupled design of sparse DiT with dynamic Mixture-of-Experts (MoE) architecture, in which two separate encoders are first involved to independently process image and text tokens. Then, a single-stream sparse DiT structure with dynamic MoE architecture is adopted to trigger multi-model interaction for image generation in a cost-efficient manner. To support flexiable accessibility with varied model capabilities, we provide HiDream-I1 in three variants: HiDream-I1-Full, HiDream-I1-Dev, and HiDream-I1-Fast. Furthermore, we go beyond the typical text-to-image generation and remould HiDream-I1 with additional image conditions to perform precise, instruction-based editing on given images, yielding a new instruction-based image editing model namely HiDream-E1. Ultimately, by integrating text-to-image generation and instruction-based image editing, HiDream-I1 evolves to form a comprehensive image agent (HiDream-A1) capable of fully interactive image creation and refinement. To accelerate multi-modal AIGC research, we have open-sourced all the codes and model weights of HiDream-I1-Full, HiDream-I1-Dev, HiDream-I1-Fast, HiDream-E1 through our project websites: https://github.com/HiDream-ai/HiDream-I1 and https://github.com/HiDream-ai/HiDream-E1. All features can be directly experienced via https://vivago.ai/studio.

cs.CV↗

arXiv2025

Diffusion Augmented Complex Maximum Total Correntropy Algorithm for Power System Frequency Estimation

Currently, adaptive filtering algorithms have been widely applied in frequency estimation for power systems. However, research on diffusion tasks remains insufficient. Existing diffusion adaptive frequency estimation algorithms exhibit certain limitations in handling input noise and lack robustness against impulsive noise. Moreover, traditional adaptive filtering algorithms designed based on the strictly-linear (SL) model fail to effectively address frequency estimation challenges in unbalanced three-phase power systems. To address these issues, this letter proposes an improved diffusion augmented complex maximum total correntropy (DAMTCC) algorithm based on the widely linear (WL) model. The proposed algorithm not only significantly enhances the capability to handle input noise but also demonstrates superior robustness to impulsive noise. Furthermore, it successfully resolves the critical challenge of frequency estimation in unbalanced three-phase power systems, offering an efficient and reliable solution for diffusion power system frequency estimation. Finally, we analyze the stability of the algorithm and computer simulations verify the excellent performance of the algorithm.

eess.SP↗

arXiv2025

Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesis

Recent advances in text-based large language models (LLMs), particularly in the GPT series and the o1 model, have demonstrated the effectiveness of scaling both training-time and inference-time compute. However, current state-of-the-art TTS systems leveraging LLMs are often multi-stage, requiring separate models (e.g., diffusion models after LLM), complicating the decision of whether to scale a particular model during training or testing. This work makes the following contributions: First, we explore the scaling of train-time and inference-time compute for speech synthesis. Second, we propose a simple framework Llasa for speech synthesis that employs a single-layer vector quantizer (VQ) codec and a single Transformer architecture to fully align with standard LLMs such as Llama. Our experiments reveal that scaling train-time compute for Llasa consistently improves the naturalness of synthesized speech and enables the generation of more complex and accurate prosody patterns. Furthermore, from the perspective of scaling inference-time compute, we employ speech understanding models as verifiers during the search, finding that scaling inference-time compute shifts the sampling modes toward the preferences of specific verifiers, thereby improving emotional expressiveness, timbre consistency, and content accuracy. In addition, we released the checkpoint and training code for our TTS model (1B, 3B, 8B) and codec model publicly available.

eess.AS↗

arXiv2024

One Pot Synthesis of Cubic Gauche Polymeric Nitrogen

The long sought cubic gauche polymeric nitrogen (cg-N) consisting of N-N single bonds has been synthesized by a simple route using sodium azide as a precursor at ambient conditions. The recrystallization process was designed to expose crystal faces with low activation energy that facilitates initiating the polymeric reaction at ambient conditions. The azide was considered as a precursor due to the low energy barrier in transforming double bonded N=N to single bonded cg-N. Raman spectrum measurements detected the emerging vibron peaks at 635 cm-1 for the polymerized sodium azide samples, demonstrating the formation of cg-N with N-N single bonds. Different from traditional high pressure technique and recently developed plasma enhanced chemical vapor deposition method, the route achieves the quantitative synthesis of cg-N at ambient conditions. The simple method to synthesize cg-N offers potential for further scale up production as well as practical applications of polymeric nitrogen based materials as high energy density materials.

cond-mat.mtrl-sci↗

arXiv2024

Dissipation-assisted preparation of topological boundary states

Robust states emerging at the boundaries of a system are an important hallmark of topological matter. Here, using the Su-Schrieffer-Heeger model and the Kitaev chain as examples, we study the impact of a type of experimentally realizable bond dissipation on topological systems by calculating the steady-state density matrix, and demonstrate that such dissipation applied near the system boundary can assist in preparing topological edge states of the parent Hamiltonian, irrespective of the initial state or filling. This effect stems from the matching between the phase distribution encoded in the topological edge states and the target state prepared through bond dissipation. This work provides new insights into the preparation of topological edge states, particularly in the context of Majorana zero modes.

cond-mat.mes-hall↗

arXiv2024

Well-posedness for the semilinear wave equations with nonlinear damping on manifolds with conical degeneration

This paper deals with a class of semilinear wave equation with nonlinear damping term $|u_{t}|^{m-2}u_t $ and nonlinear source term $g(x)|u|^{p-2}u$ on the manifolds with conical singularities. Firstly, we prove the local existence and uniqueness of the solution by the semigroup method. Secondly, we establish the global existence, the energy decay estimate and the blow-up in finite time of the solution with subcritial ($E(0)<d$) and critial ($E(0)=d$) initial energy level by constructing potential wells. We also show that the solution is global provided the damping dominates the source (that is $m\geq p$). Moreover, we prove the blow-up in finite time of the solution with arbitrary positive initial energy and give some necessary and sufficient condition for existing finite time blow-up solutions.

math.AP↗

arXiv2024

A facile route to synthesize cubic gauche polymeric nitrogen

In this work, the long-sought cg-N with N-N single bond has been synthesized for the first time by a thermal-driven-only chemical route at ambient conditions. The successful synthesis of cg-N was achieved by first creating a solution of azides, which was then pretreated under vacuum conditions. Following the pretreatment, the resultant concentrated azide was heated at temperatures ranging from 260°C to 330°C for a reaction time of 3 hours, ultimately leading to the formation of cg-N. The emergent intense Raman peak characterized of cg-N provides solid evidence that the double bonded nitrogen-nitrogen transforms into a single bond form, which agrees well with cg-N structure. To date, this is the only work achieving the quantity of cg-N synthesized at ambient conditions by a facile route that can be further developed for the scalable synthesis and applications of polymerized nitrogen-based materials as high energy density materials.

cond-mat.mtrl-sci↗