Searcharxiv⌕ Search

arXiv subjects

Kai Zhu

Publications and source records attributed to Kai Zhu.

At least 37 records · Page 2Linked to original sources
arXiv2026

TaCauchy: An Extensible FEM Framework for Vision-Based Tactile Simulation

Vision-based tactile sensors require high-fidelity simulation for reinforcement learning, yet existing approaches struggle to provide accurate mechanical stress fields within GPU-accelerated robotics platforms. We present TaCauchy, an extensible Finite Element Method (FEM) framework that integrates rigorous physics-based force computation into Isaac Sim. Built on the Unified Incremental Potential Contact (UIPC) solver, TaCauchy directly computes Cauchy stress tensors from hyperelastic constitutive laws and projects them onto contact surfaces to obtain traction forces and pressure distributions, providing mechanical ground truth from first principles rather than empirical estimation. Our framework features automatic mesh generation with geometry-aware adaptive refinement and a modular sensor interface enabling rapid integration of diverse sensors (GelSight Mini, DIGIT, 9DTact) with minimal configuration. Performance benchmarks demonstrate 33.40 FPS for single environments and 555 FPS aggregate throughput across 60 parallel environments, with stress extraction overhead under 1 ms. Physical validation experiments show strong agreement between simulated and real tactile responses across force ranges from 1.2556 N to 4.7332 N, achieving SSIM above 0.93, confirming the framework's capability to provide accurate, physically-grounded force supervision for downstream robotic manipulation tasks.

cs.RO↗
arXiv2026

CUTh-Solver: GPU-Accelerated Sparse Matrix Solver for High-Resolution Thermal Simulation of 3D ICs

Coarse-grained thermal simulation tends to underestimate localized thermal issues, potentially missing critical hotspots. Accurate analysis, therefore, demands fine-grained information, which dramatically increases grid resolution and thus computational workload. Fortunately, the coefficient matrices are often sparse with regular sparsity patterns, offering optimization opportunities. However, existing general-purpose matrix solvers on GPUs rarely exploit these domain-specific properties, thereby encountering bottlenecks in data storage, memory access, parallelism, computational efficiency, and hardware utilization. Therefore, we propose CUTh-Solver, a co-designed GPU-accelerated Preconditioned Conjugate Gradient (PCG)-based sparse solver framework for Symmetric Positive Definite (SPD) systems arising from high-resolution steady-state and transient 3D IC thermal simulation. For data storage, CUTh-Solver condenses the Diagonal (DIA) storage format to remove redundancy. To optimize the memory access, CUTh-Solver employs diagonal-wise SpMV to achieve coalesced memory access. We further observe a critical conflict between parallelism and preconditioning quality and thus adopt a high-parallelism preconditioning strategy. To improve computational efficiency and hardware utilization, we employ an adaptive fine-grained mixed-precision strategy that leverages diverse floating-point units to avoid resource contention, enhancing throughput without compromising numerical stability. Experimental results show that CUTh-Solver achieves up to 25.8x speedup over GPU-accelerated COMSOL Multiphysics 6.4 and over 3x speedup over NVIDIA's native general-purpose libraries (AmgX, cuSPARSE, cuDSS). Ablation studies validate the individual contribution of each optimization. The code is available at: https://github.com/Chenghan-Wang/CUTh-Solver

cs.AR↗
arXiv2026

A Strong Stellar Age-Metallicity Gradient Relation in Nearby Dwarf Galaxies Driven by Stellar Migration and Environmental Quenching

Stellar metallicity gradients ($\nabla[Z/H]$) provide a fossil record of the assembly history of galaxies. We present an analysis of $\nabla[Z/H]$ for 90 nearby low-mass galaxies using VLT/MUSE IFU spectroscopy, spanning stellar masses from $10^{6.5}$ to $10^{10} M_\odot$ (median $\sim 10^{8.5} M_\odot$) and significantly extending the mass coverage of existing IFU surveys into the classical dwarf regime. Our primary finding is a robust negative correlation between $\nabla[Z/H]$ and light-weighted stellar age ($|r|\gtrsim 0.7$) measured out to $\sim$ 2$\times$ effective radius: older dwarf galaxies have steeper (more negative) gradients. This holds regardless of stellar mass, structural compactness, or large-scale environment (group/field), and is strongest in the intermediate-mass regime ($8.2\lesssim\log M_\star/M_\odot\lesssim9.0$). The slope of the age-$\nabla[Z/H]$ relation is close to that in the FIRE-2 simulations, indicating that stellar radial migration driven by feedback-induced potential fluctuations may be fundamental in dwarf evolution. But this apparent consistency is likely coincidental given the simulations' overly efficient feedback and chemical mixing. On the other hand, the H\,\textsc{i} deficiency parameter, an indicator of past environmental stripping, shows a moderate yet highly significant correlation with $\nabla[Z/H]$, second only to stellar age in strength: galaxies with higher H\,\textsc{i} deficiency tend to have more negative gradients, strongly indicating that environment-driven outside-in quenching and the ensuing gradual truncation of metal enrichment re-shape the stellar metallicity distribution. Our analysis suggests that the chemical evolution of dwarf galaxies likely arises from a synergy of feedback-driven dynamical heating and external environmental processing, though only the latter has robust observational support.

astro-ph.GA↗
arXiv2026

Wan-Image: Pushing the Boundaries of Generative Visual Intelligence

We present Wan-Image, a unified visual generation system explicitly engineered to paradigm-shift image generation models from casual synthesizers into professional-grade productivity tools. While contemporary diffusion models excel at aesthetic generation, they frequently encounter critical bottlenecks in rigorous design workflows that demand absolute controllability, complex typography rendering, and strict identity preservation. To address these challenges, Wan-Image features a natively unified multi-modal architecture by synergizing the cognitive capabilities of large language models with the high-fidelity pixel synthesis of diffusion transformers, which seamlessly translates highly nuanced user intents into precise visual outputs. It is fundamentally powered by large-scale multi-modal data scaling, a systematic fine-grained annotation engine, and curated reinforcement learning data to surpass basic instruction following and unlock expert-level professional capabilities. These include ultra-long complex text rendering, hyper-diverse portrait generation, palette-guided generation, multi-subject identity preservation, coherent sequential visual generation, precise multi-modal interactive editing, native alpha-channel generation, and high-efficiency 4K synthesis. Across diverse human evaluations, Wan-Image exceeds Seedream 5.0 Lite and GPT Image 1.5 in overall performance, reaching parity with Nano Banana Pro in challenging tasks. Ultimately, Wan-Image revolutionizes visual content creation across e-commerce, entertainment, education, and personal productivity, redefining the boundaries of professional visual synthesis.

cs.CV↗
arXiv2026

RoboCOIN: An Open-Sourced Bimanual Robotic Data Collection for Integrated Manipulation

Despite the critical role of bimanual manipulation in endowing robots with human-like dexterity, large-scale and diverse datasets remain scarce due to the significant hardware heterogeneity across bimanual robotic platforms. To bridge this gap, we introduce RoboCOIN, a large-scale multi-embodiment bimanual manipulation dataset comprising over 180,000 demonstrations collected from 15 distinct robotic platforms. Spanning 16 diverse environments-including residential, commercial, and industrial settings-the dataset features 421 bimanual tasks systematically categorized by 39 bimanual collaboration actions and 432 objects. A key innovation of our work is the hierarchical capability pyramid, which provides granular annotations ranging from trajectory-level concepts to segment-level subtasks and frame-level kinematics. Furthermore, we present CoRobot, an efficient data processing pipeline powered by the Robot Trajectory Markup Language (RTML), designed to facilitate quality assessment, automated annotation, and unified multi-embodiment and data management. Extensive experiments demonstrate the effectiveness of RoboCOIN in enhancing the performance of various bimanual manipulation models across a wide spectrum of robotic embodiments. The entire dataset and codebase are fully open-sourced, providing a valuable resource for advancing research in bimanual and multi-embodiment manipulation.

cs.RO↗
arXiv2026

The Role of Baryonic and Dark Matter in Bar Kinematics

Simulations predict that bars in galaxies should slow down over time. This is often attributed to the exchange of angular momentum between the bar and other regions of the galaxy, such as the outer disc and dark matter halo, which implies that galaxies with a more massive halo or disc should be able to slow down the bar more efficiently. However, observational evidence for this process has been limited. In this work, we provide observational support for the slowing down of bars as predicted by simulations. We combine bar kinematics measurements obtained with the Tremaine-Weinberg method and host galaxy mass estimates derived from Jeans anisotropic models for a sample of 30 galaxies from the MaNGA survey. We find a statistically significant anti-correlation (>4sigma) between the bar pattern speed and both the stellar and total dynamical mass, which suggests that the slowest bars reside in the most massive galaxies. However, while the slope of the best-fit line between the pattern speed and dark matter mass is negative, it is not statistically significant (2.43sigma). We also find that bars with lower pattern speeds have more extended NFW dark matter profiles with lower central densities. Additionally, we find statistically significant correlations (>3sigma) between the corotation radius and the stellar mass, dark matter mass, and total dynamical mass. Finally, we find no significant correlations that involve the dark matter fraction or R, likely due to the inherent challenges associated with measuring these specific parameters accurately.

astro-ph.GA↗
arXiv2026

Wan-Weaver: Interleaved Multi-modal Generation via Decoupled Training

Recent unified models have made unprecedented progress in both understanding and generation. However, while most of them accept multi-modal inputs, they typically produce only single-modality outputs. This challenge of producing interleaved content is mainly due to training data scarcity and the difficulty of modeling long-range cross-modal context. To address this issue, we decompose interleaved generation into textual planning and visual consistency modeling, and introduce a framework consisting of a planner and a visualizer. The planner produces dense textual descriptions for visual content, while the visualizer synthesizes images accordingly. Under this guidance, we construct large-scale textual-proxy interleaved data (where visual content is represented in text) to train the planner, and curate reference-guided image data to train the visualizer. These designs give rise to Wan-Weaver, which exhibits emergent interleaved generation ability with long-range textual coherence and visual consistency. Meanwhile, the integration of diverse understanding and generation data into planner training enables Wan-Weaver to achieve robust task reasoning and generation proficiency. To assess the model's capability in interleaved generation, we further construct a benchmark that spans a wide range of use cases across multiple dimensions. Extensive experiments demonstrate that, even without access to any real interleaved data, Wan-Weaver achieves superior performance over existing methods.

cs.CV↗
arXiv2026

RS-SSM: Refining Forgotten Specifics in State Space Model for Video Semantic Segmentation

Recently, state space models have demonstrated efficient video segmentation through linear-complexity state space compression. However, Video Semantic Segmentation (VSS) requires pixel-level spatiotemporal modeling capabilities to maintain temporal consistency in segmentation of semantic objects. While state space models can preserve common semantic information during state space compression, the fixed-size state space inevitably forgets specific information, which limits the models' capability for pixel-level segmentation. To tackle the above issue, we proposed a Refining Specifics State Space Model approach (RS-SSM) for video semantic segmentation, which performs complementary refining of forgotten spatiotemporal specifics. Specifically, a Channel-wise Amplitude Perceptron (CwAP) is designed to extract and align the distribution characteristics of specific information in the state space. Besides, a Forgetting Gate Information Refiner (FGIR) is proposed to adaptively invert and refine the forgetting gate matrix in the state space model based on the specific information distribution. Consequently, our RS-SSM leverages the inverted forgetting gate to complementarily refine the specific information forgotten during state space compression, thereby enhancing the model's capability for spatiotemporal pixel-level segmentation. Extensive experiments on four VSS benchmarks demonstrate that our RS-SSM achieves state-of-the-art performance while maintaining high computational efficiency. The code is available at https://github.com/zhoujiahuan1991/CVPR2026-RS-SSM.

cs.CV↗
arXiv2026

DepthLab: From Partial to Complete

Missing values remain a common challenge for depth data across its wide range of applications, stemming from various causes like incomplete data acquisition and perspective alteration. This work bridges this gap with DepthLab, a foundation depth inpainting model powered by image diffusion priors. Our model features two notable strengths: (1) it demonstrates resilience to depth-deficient regions, providing reliable completion for both continuous areas and isolated points, and (2) it faithfully preserves scale consistency with the conditioned known depth when filling in missing values. Drawing on these advantages, our approach proves its worth in various downstream tasks, including 3D scene inpainting, text-to-3D scene generation, sparse-view reconstruction with DUST3R, and LiDAR depth completion, exceeding current solutions in both numerical performance and visual quality. Our project page with source code is available at https://johanan528.github.io/depthlab_web/.

cs.CV↗
arXiv2026

SimRecon: SimReady Compositional Scene Reconstruction from Real Videos

Compositional scene reconstruction seeks to create object-centric representations rather than holistic scenes from real-world videos, which is natively applicable for simulation and interaction. Conventional compositional reconstruction approaches primarily emphasize on visual appearance and show limited generalization ability to real-world scenarios. In this paper, we propose SimRecon, a framework that realizes a "Perception-Generation-Simulation" pipeline towards cluttered scene reconstruction, which first conducts scene-level semantic reconstruction from video input, then performs single-object generation, and finally assembles these assets in the simulator. However, naively combining these three stages leads to visual infidelity of generated assets and physical implausibility of the final scene, a problem particularly severe for complex scenes. Thus, we further propose two bridging modules between the three stages to address this problem. To be specific, for the transition from Perception to Generation, critical for visual fidelity, we introduce Active Viewpoint Optimization, which actively searches in 3D space to acquire optimal projected images as conditions for single-object completion. Moreover, for the transition from Generation to Simulation, essential for physical plausibility, we propose a Scene Graph Synthesizer, which guides the construction from scratch in 3D simulators, mirroring the native, constructive principle of the real world. Extensive experiments on the ScanNet dataset validate our method's superior performance over previous state-of-the-art approaches.

cs.CV↗
arXiv2026

AoE: Always-on Egocentric Human Video Collection for Embodied AI

Embodied foundation models require large-scale, high-quality real-world interaction data for pre-training and scaling. However, existing data collection methods suffer from high infrastructure costs, complex hardware dependencies, and limited interaction scope, making scalable expansion challenging. In fact, humans themselves are ideal physically embodied agents. Therefore, obtaining egocentric real-world interaction data from globally distributed "human agents" offers advantages of low cost and sustainability. To this end, we propose the Always-on Egocentric (AoE) data collection system, which aims to simplify hardware dependencies by leveraging humans themselves and their smartphones, enabling low-cost, highly efficient, and scene-agnostic real-world interaction data collection to address the challenge of data scarcity. Specifically, we first employ an ergonomic neck-mounted smartphone holder to enable low-barrier, large-scale egocentric data collection through a cloud-edge collaborative architecture. Second, we develop a cross-platform mobile APP that leverages on-device compute for real-time processing, while the cloud hosts automated labeling and filtering pipelines that transform raw videos into high-quality training data. Finally, the AoE system supports distributed Ego video data collection by anyone, anytime, and anywhere. We evaluate AoE on data preprocessing quality and downstream tasks, demonstrating that high-quality egocentric data significantly boosts real-world generalization.

cs.CV↗
arXiv2026

Towards Sequence Modeling Alignment between Tokenizer and Autoregressive Model

Autoregressive image generation aims to predict the next token based on previous ones. However, this process is challenged by the bidirectional dependencies inherent in conventional image tokenizations, which creates a fundamental misalignment with the unidirectional nature of autoregressive models. To resolve this, we introduce AliTok, a novel Aligned Tokenizer that alters the dependency structure of the token sequence. AliTok employs a bidirectional encoder constrained by a causal decoder, a design that compels the encoder to produce a token sequence with both semantic richness and forward-dependency. Furthermore, by incorporating prefix tokens and employing a two-stage tokenizer training process to enhance reconstruction performance, AliTok achieves high fidelity and predictability simultaneously. Building upon AliTok, a standard decoder-only autoregressive model with just 177M parameters achieves a gFID of 1.44 and an IS of 319.5 on ImageNet-256. Scaling to 662M, our model reaches a gFID of 1.28, surpassing the SOTA diffusion method with 10x faster sampling. On ImageNet-512, our 318M model also achieves a SOTA gFID of 1.39. Code and weights at https://github.com/ali-vilab/alitok.

cs.CV↗
arXiv2026

Unifying the dynamical classification of early-type galaxies: kinematic deficits in IllustrisTNG versus observations

We conduct a comparative analysis of galaxy kinematics using IllustrisTNG simulations and integral-field spectroscopy (IFS) observations. We identify 2,342 early-type galaxies (ETGs) from the TNG100 simulation and 236 ETGs from the TNG50 simulation, comparing them with observations from MaNGA and ATLAS$^{3D}$. For these systems, we measure key kinematic parameters: the intrinsic spin parameter $λ_{R,\mathrm{intr}}$ (measured edge-on), the cylindrical rotational energy fraction $κ_{\mathrm{rot}}$, and structural mass ratios including the spheroid mass fraction $f_{\mathrm{spheroid}}$ and stellar halo mass fraction $f_{\mathrm{halo}}$. Our study reveals that standard classifiers--the $λ_{R}(R_e)=0.31\sqrt{\varepsilon}$ relation and $\overline{k_5}$ coefficient (higher-order Fourier term of velocity fields)--fail to align with observed kinematic bimodality. We propose revised thresholds: $λ_{R,\mathrm{intr}} \sim 0.4$, $κ_{\mathrm{rot}} \sim 0.5$, and $f_{\mathrm{spheroid}} \sim 0.6$, which classify galaxies into rotation-dominated (fast rotators) and random motion-dominated (slow rotators). Scaling relations from TNG enable observational estimates of $κ_{\mathrm{rot}}$ and $f_{\mathrm{spheroid}}$. The simulations exhibit a bimodality deficit, characterized by a lack of fast rotators and suppressed $λ_{R,\mathrm{intr}}$, attributed to excess galaxies with intermediate rotation and high spheroid/stellar halo mass. We introduce a novel method to estimate $f_{\mathrm{halo}}$ from IFS kinematics, though uncertainties remain.

astro-ph.GA↗
arXiv2026

Collaborative Belief Reasoning with LLMs for Efficient Multi-Agent Collaboration

Effective real-world multi-agent collaboration requires not only accurate planning but also the ability to reason about collaborators' intents--a crucial capability for avoiding miscoordination and redundant communication under partial observable environments. Due to their strong planning and reasoning capabilities, large language models (LLMs) have emerged as promising autonomous agents for collaborative task solving. However, existing collaboration frameworks for LLMs overlook their reasoning potential for dynamic intent inference, and thus produce inconsistent plans and redundant communication, reducing collaboration efficiency. To bridge this gap, we propose CoBel-World, a novel framework that equips LLM agents with a Collaborative Belief World--an internal representation jointly modeling the physical environment and collaborators' mental states. CoBel-World enables agents to parse external open-world knowledge into structured beliefs via a symbolic belief representation module, and perform zero-shot Bayesian-style belief updates through LLM reasoning. This allows agents to proactively detect potential miscoordination (e.g., conflicting plans) and communicate adaptively. Evaluated on challenging embodied benchmarks (i.e., TDW-MAT and C-WAH), CoBel-World significantly reduces communication costs by 64-79% and improves task completion efficiency by 4-28% compared to the strongest baseline. Our results show that explicit, intent-aware belief modeling is essential for efficient and human-like collaboration in LLM-based multi-agent systems.

cs.AI↗
arXiv2026

EditEmoTalk: Controllable Speech-Driven 3D Facial Animation with Continuous Expression Editing

Speech-driven 3D facial animation aims to generate realistic and expressive facial motions directly from audio. While recent methods achieve high-quality lip synchronization, they often rely on discrete emotion categories, limiting continuous and fine-grained emotional control. We present EditEmoTalk, a controllable speech-driven 3D facial animation framework with continuous emotion editing. The key idea is a boundary-aware semantic embedding that learns the normal directions of inter-emotion decision boundaries, enabling a continuous expression manifold for smooth emotion manipulation. Moreover, we introduce an emotional consistency loss that enforces semantic alignment between the generated motion dynamics and the target emotion embedding through a mapping network, ensuring faithful emotional expression. Extensive experiments demonstrate that EditEmoTalk achieves superior controllability, expressiveness, and generalization while maintaining accurate lip synchronization. Code and pretrained models will be released.

cs.MM↗
arXiv2025

UCS: A Universal Model for Curvilinear Structure Segmentation

Curvilinear structure segmentation (CSS) is essential in various domains, including medical imaging, landscape analysis, industrial surface inspection, and plant analysis. While existing methods achieve high performance within specific domains, their generalizability is limited. On the other hand, large-scale models such as Segment Anything Model (SAM) exhibit strong generalization but are not optimized for curvilinear structures. Existing adaptations of SAM primarily focus on general object segmentation and lack specialized design for CSS tasks. To bridge this gap, we propose the Universal Curvilinear structure Segmentation (UCS) model, which adapts SAM to CSS tasks while further enhancing its cross-domain generalization. UCS features a novel encoder architecture integrating a pretrained SAM encoder with two innovations: a Sparse Adapter, strategically inserted to inherit the pre-trained SAM encoder's generalization capability while minimizing the number of fine-tuning parameters, and a Prompt Generation module, which leverages Fast Fourier Transform with a high-pass filter to generate curve-specific prompts. Furthermore, the UCS incorporates a mask decoder that eliminates reliance on manual interaction through a dual-compression module: a Hierarchical Feature Compression module, which aggregates the outputs of the sampled encoder to enhance detail preservation, and a Guidance Feature Compression module, which extracts and compresses image-driven guidance features. Evaluated on a comprehensive multi-domain dataset, including an in-house dataset covering eight natural curvilinear structures, UCS demonstrates state-of-the-art generalization and open-set segmentation performance across medical, engineering, natural, and plant imagery, establishing a new benchmark for universal CSS. The source code is available at https://github.com/kylechuuuuu/UCS.

cs.CV↗
arXiv2025

FACM: Flow-Anchored Consistency Models

Continuous-time Consistency Models (CMs) promise efficient few-step generation but face significant challenges with training instability. We argue this instability stems from a fundamental conflict: Training the network exclusively on a shortcut objective leads to the catastrophic forgetting of the instantaneous velocity field that defines the flow. Our solution is to explicitly anchor the model in the underlying flow, ensuring high trajectory fidelity during training. We introduce the Flow-Anchored Consistency Model (FACM), where a Flow Matching (FM) task serves as a dynamic anchor for the primary CM shortcut objective. Key to this Flow-Anchoring approach is a novel expanded time interval strategy that unifies optimization for a single model while decoupling the two tasks to ensure stable, architecturally-agnostic training. By distilling a pre-trained LightningDiT model, our method achieves a state-of-the-art FID of 1.32 with two steps (NFE=2) and 1.70 with just one step (NFE=1) on ImageNet 256x256. To address the challenge of scalability, we develop a memory-efficient Chain-JVP that resolves key incompatibilities with FSDP. This method allows us to scale FACM training on a 14B parameter model (Wan 2.2), accelerating its Text-to-Image inference from 2x40 to 2-8 steps. Our code and pretrained models: https://github.com/ali-vilab/FACM.

cs.CV↗
arXiv2025

Anchoring Values in Temporal and Group Dimensions for Flow Matching Model Alignment

Group Relative Policy Optimization (GRPO) has proven highly effective in enhancing the alignment capabilities of Large Language Models (LLMs). However, current adaptations of GRPO for the flow matching-based image generation neglect a foundational conflict between its core principles and the distinct dynamics of the visual synthesis process. This mismatch leads to two key limitations: (i) Uniformly applying a sparse terminal reward across all timesteps impairs temporal credit assignment, ignoring the differing criticality of generation phases from early structure formation to late-stage tuning. (ii) Exclusive reliance on relative, intra-group rewards causes the optimization signal to fade as training converges, leading to the optimization stagnation when reward diversity is entirely depleted. To address these limitations, we propose Value-Anchored Group Policy Optimization (VGPO), a framework that redefines value estimation across both temporal and group dimensions. Specifically, VGPO transforms the sparse terminal reward into dense, process-aware value estimates, enabling precise credit assignment by modeling the expected cumulative reward at each generative stage. Furthermore, VGPO replaces standard group normalization with a novel process enhanced by absolute values to maintain a stable optimization signal even as reward diversity declines. Extensive experiments on three benchmarks demonstrate that VGPO achieves state-of-the-art image quality while simultaneously improving task-specific accuracy, effectively mitigating reward hacking. Project webpage: https://yawen-shao.github.io/VGPO/.

cs.LG↗