arXiv subjects

Qi Wang

Publications and source records attributed to Qi Wang.

At least 217 records · Page 12Linked to original sources

arXiv2025

Dual-Actor Fine-Tuning of VLA Models: A Talk-and-Tweak Human-in-the-Loop Approach

Vision-language-action (VLA) models demonstrate strong generalization in robotic manipulation but face challenges in complex, real-world tasks. While supervised fine-tuning with demonstrations is constrained by data quality, reinforcement learning (RL) offers a promising alternative. We propose a human-in-the-loop dual-actor fine-tuning framework grounded in RL. The framework integrates a primary actor for robust multi-task performance with a refinement actor for latent-space adaptation. Beyond standard physical interventions, we introduce a lightweight talk-and-tweak scheme that converts human corrections into semantically grounded language commands, thereby generating a new dataset for policy learning. In real-world multi-task experiments, our approach achieves 100% success across three tasks within 101 minutes of online fine-tuning. For long-horizon tasks, it sustains a 50% success rate over 12 consecutive operations. Furthermore, the framework scales effectively to multi-robot training, achieving up to a 2 times improvement in efficiency when using dual robots. The experiment videos are available at https://sites.google.com/view/hil-daft/.

cs.RO↗

arXiv2025

Wan-Animate: Unified Character Animation and Replacement with Holistic Replication

We introduce Wan-Animate, a unified framework for character animation and replacement. Given a character image and a reference video, Wan-Animate can animate the character by precisely replicating the expressions and movements of the character in the video to generate high-fidelity character videos. Alternatively, it can integrate the animated character into the reference video to replace the original character, replicating the scene's lighting and color tone to achieve seamless environmental integration. Wan-Animate is built upon the Wan model. To adapt it for character animation tasks, we employ a modified input paradigm to differentiate between reference conditions and regions for generation. This design unifies multiple tasks into a common symbolic representation. We use spatially-aligned skeleton signals to replicate body motion and implicit facial features extracted from source images to reenact expressions, enabling the generation of character videos with high controllability and expressiveness. Furthermore, to enhance environmental integration during character replacement, we develop an auxiliary Relighting LoRA. This module preserves the character's appearance consistency while applying the appropriate environmental lighting and color tone. Experimental results demonstrate that Wan-Animate achieves state-of-the-art performance. We are committed to open-sourcing the model weights and its source code.

cs.CV↗

arXiv2025

Adaptive Sampling Scheduler

Consistent distillation methods have evolved into effective techniques that significantly accelerate the sampling process of diffusion models. Although existing methods have achieved remarkable results, the selection of target timesteps during distillation mainly relies on deterministic or stochastic strategies, which often require sampling schedulers to be designed specifically for different distillation processes. Moreover, this pattern severely limits flexibility, thereby restricting the full sampling potential of diffusion models in practical applications. To overcome these limitations, this paper proposes an adaptive sampling scheduler that is applicable to various consistency distillation frameworks. The scheduler introduces three innovative strategies: (i) dynamic target timestep selection, which adapts to different consistency distillation frameworks by selecting timesteps based on their computed importance; (ii) Optimized alternating sampling along the solution trajectory by guiding forward denoising and backward noise addition based on the proposed time step importance, enabling more effective exploration of the solution space to enhance generation performance; and (iii) Utilization of smoothing clipping and color balancing techniques to achieve stable and high-quality generation results at high guidance scales, thereby expanding the applicability of consistency distillation models in complex generation scenarios. We validated the effectiveness and flexibility of the adaptive sampling scheduler across various consistency distillation methods through comprehensive experimental evaluations. Experimental results consistently demonstrated significant improvements in generative performance, highlighting the strong adaptability achieved by our method.

cs.CV↗

arXiv2025

Global Bifurcations and Pattern Formation in Target-Offender-Guardian Crime Models

We study a reaction-advection-diffusion model of a target-offender-guardian system designed to capture interactions between urban crime and policing. Using Crandall-Rabinowitz bifurcation theory and spectral analysis, we establish rigorous conditions for both steady-state and Hopf bifurcations. These results identify critical thresholds of policing intensity at which spatially uniform equilibria lose stability, leading either to persistent heterogeneous hotspots or oscillatory crime-policing cycles. From a criminological perspective, such thresholds represent tipping points in guardian mobility: once crossed, they can lock neighborhoods into stable clusters of criminal activity or trigger recurrent waves of hotspot formation. Numerical simulations complement the theory, exhibiting stationary patterns, periodic oscillations, and chaotic dynamics. By explicitly incorporating law enforcement as a third interacting component, our framework extends classical two-equation models. It offers new tools for analyzing nonlinear interactions, bifurcations, and pattern formation in multi-agent social systems.

nlin.PS↗

arXiv2025

GFS: A Preemption-aware Scheduling Framework for GPU Clusters with Predictive Spot Instance Management

The surge in large language models (LLMs) has fundamentally reshaped the landscape of GPU usage patterns, creating an urgent need for more efficient management strategies. While cloud providers employ spot instances to reduce costs for low-priority (LP) tasks, existing schedulers still grapple with high eviction rates and lengthy queuing times. To address these limitations, we present GFS, a novel preemptive scheduling framework that enhances service-level objective (SLO) compliance for high-priority (HP) tasks while minimizing preemptions to LP tasks. Firstly, GFS utilizes a lightweight forecasting model that predicts GPU demand among different tenants, enabling proactive resource management. Secondly, GFS employs a dynamic allocation mechanism to adjust the spot quota for LP tasks with guaranteed durations. Lastly, GFS incorporates a preemptive scheduling policy that prioritizes HP tasks while minimizing the impact on LP tasks. We demonstrate the effectiveness of GFS through both real-world implementation and simulations. The results show that GFS reduces eviction rates by 33.0\%, and cuts queuing delays by 44.1\% for LP tasks. Furthermore, GFS enhances the GPU allocation rate by up to 22.8\% in real production clusters. In a production cluster of more than 10,000 GPUs, GFS yields roughly \$459,715 in monthly benefits.

cs.DC↗

arXiv2025

Data-Driven Discovery of Mobility Periodicity for Understanding Urban Systems

Human mobility regularity is crucial for understanding urban dynamics and informing decision-making processes. This study first quantifies the periodicity in complex human mobility data as a sparse identification of dominant positive auto-correlations in time series autoregression and then discovers periodic patterns. We apply the framework to large-scale metro passenger flow data in Hangzhou, China and multi-modal mobility data in New York City and Chicago, USA, revealing the interpretable weekly periodicity across different spatial locations over past several years. The analysis of ridesharing data from 2019 to 2024 demonstrates the disruptive impact of the pandemic on mobility regularity and the subsequent recovery trends. In 2024, the periodic mobility patterns of ridesharing, taxi, subway, and bikesharing in Manhattan uncover the regularity and variability of these travel modes. Our findings highlight the potential of interpretable machine learning to discover spatiotemporal mobility patterns and offer a valuable tool for understanding urban systems.

cs.SI↗

arXiv2025

Klear-CodeTest: Scalable Test Case Generation for Code Reinforcement Learning

Precise, correct feedback is crucial for effectively training large language models (LLMs) in code reinforcement learning. However, synthesizing high-quality test cases remains a profoundly challenging and unsolved problem. In this work, we present Klear-CodeTest, a comprehensive test case synthesis framework featuring rigorous verification to ensure quality and reliability of test cases. Our approach achieves broad coverage of programming problems via a novel Generator-Validation (G-V) framework, ensuring correctness through a consistency validation mechanism that verifies outputs against gold solutions. The proposed G-V framework generates comprehensive test cases including both regular and corner cases, enhancing test coverage and discriminative power for solution correctness assessment in code reinforcement learning. In addition, we design a multi-layered security sandbox system optimized for online verification platforms, guaranteeing safe and reliable code execution. Through comprehensive experiments, we demonstrate the effectiveness of our curated dataset, showing significant improvements in model performance and training stability. The source codes, curated dataset and sandbox system are available at: https://github.com/Kwai-Klear/CodeTest.

cs.SE↗

arXiv2025

MimicDroid: In-Context Learning for Humanoid Robot Manipulation from Human Play Videos

We aim to enable humanoid robots to efficiently solve new manipulation tasks from a few video examples. In-context learning (ICL) is a promising framework for achieving this goal due to its test-time data efficiency and rapid adaptability. However, current ICL methods rely on labor-intensive teleoperated data for training, which restricts scalability. We propose using human play videos -- continuous, unlabeled videos of people interacting freely with their environment -- as a scalable and diverse training data source. We introduce MimicDroid, which enables humanoids to perform ICL using human play videos as the only training data. MimicDroid extracts trajectory pairs with similar manipulation behaviors and trains the policy to predict the actions of one trajectory conditioned on the other. Through this process, the model acquired ICL capabilities for adapting to novel objects and environments at test time. To bridge the embodiment gap, MimicDroid first retargets human wrist poses estimated from RGB videos to the humanoid, leveraging kinematic similarity. It also applies random patch masking during training to reduce overfitting to human-specific cues and improve robustness to visual differences. To evaluate few-shot learning for humanoids, we introduce an open-source simulation benchmark with increasing levels of generalization difficulty. MimicDroid outperformed state-of-the-art methods and achieved nearly twofold higher success rates in the real world. Additional materials can be found on: ut-austin-rpl.github.io/MimicDroid

cs.RO↗

arXiv2025

Exchange spin-wave propagation in Ga:YIG nanowaveguides

Spin-wave-based computing has emerged as a promising approach to overcome the fundamental limitations of CMOS technologies. However, the increasing demand for device miniaturization down to a 100 nm scale presents significant challenges for long-distance spin-wave transport. Gallium-substituted yttrium iron garnet (Ga:YIG) offers a potential solution to these challenges due to its unique magnetic properties. The reduced saturation magnetization in Ga:YIG enables efficient excitation of exchange-dominated spin waves, which exhibit enhanced transport characteristics compared to dipolar-dominated modes in conventional materials. Here, we present the first comprehensive study combining experimental, analytical, and numerical investigations of spin-wave propagation in Ga:YIG waveguides down to 145 nm width and 73 nm thickness. Using micro-focused Brillouin light scattering spectroscopy, TetraX simulations, and analytical dispersion calculations, we demonstrate that Ga:YIG waveguides support spin waves with significantly higher group velocities up to 600 m/s. This value remains constant for structures with different widths, leading to longer spin-wave propagation lengths in nanowaveguides compared to non-substituted YIG. These results reveal that gallium substitution provides access to faster and longer-lived spin waves, opening new possibilities for implementing this material in nanoscale magnonic devices.

cond-mat.other↗

arXiv2025

A Strongly Anisotropic Superconducting Gap in the Kagome Superconductor CsV$_3$Sb$_5$: A Study of Directional Point-Contact Andreev Reflection Spectroscopy

In the recently discovered V-based kagome superconductors AV$_3$Sb$_5$ (A = K, Rb, and Cs), superconductivity is intertwined with an unconventional charge density wave (CDW) order, raising a fundamental concern on the superconducting gap structure of such kagome superconductors in the presence of CDW orders. Here, we report directional soft point-contact Andreev reflection (SPCAR) spectroscopy measurements on the kagome superconductor CsV$_3$Sb$_5$, revealing compelling evidence for the existence of a strongly anisotropic superconducting gap pairing state. The SPCAR spectra measured with current injected parallel to the $ab$-plane exhibit an in-gap single conductance peak, in contrast to those of SPCAR spectra: a double-peak structure in the perpendicular direction. These spectra are well described by an anisotropic single-gap BTK model. The extracted superconducting gaps comprise an isotropic large gap and a strongly anisotropic gap, originating from different Fermi surface sheets. Quantitative analysis reveals an anisotropy around $\ sim$70\% with a gap minimum of about 0.15 meV. These results shed new light on the unconventional multiband pairing states in kagome superconductors.

cond-mat.supr-con↗

arXiv2025

Design of RIS-UAV-Assisted LEO Satellite Constellation Communication

Low Earth orbit (LEO) satellite constellations play a pivotal role in sixth-generation (6G) wireless networks by providing global coverage, massive connections, and huge capacity. In this paper, we present a novel LEO satellite constellation communication framework, where a reconfigurable intelligent surface-mounted unmanned aerial vehicle (RIS-UAV) is deployed to improve the communication quality of multiple terrestrial user equipments (UEs) under the condition of long distance between satellite and ground. To reduce the overhead for channel state information (CSI) acquisition with multiple-satellite collaboration, statistical CSI (sCSI) is utilized in the system. In such a situation, we first derive an approximated but exact expression for ergodic rate of each UE. Then, we aim to maximize the minimum approximated UE ergodic rate by the proposed alternating optimization (AO)-based algorithm that jointly optimizes LEO satellite beamforming, RIS phase shift, and UAV trajectory. Finally, extensive simulations are conducted to demonstrate the superiority of the proposed algorithm in terms of spectrum efficiency over baseline algorithms.

cs.IT↗

arXiv2025

Wan-S2V: Audio-Driven Cinematic Video Generation

Current state-of-the-art (SOTA) methods for audio-driven character animation demonstrate promising performance for scenarios primarily involving speech and singing. However, they often fall short in more complex film and television productions, which demand sophisticated elements such as nuanced character interactions, realistic body movements, and dynamic camera work. To address this long-standing challenge of achieving film-level character animation, we propose an audio-driven model, which we refere to as Wan-S2V, built upon Wan. Our model achieves significantly enhanced expressiveness and fidelity in cinematic contexts compared to existing approaches. We conducted extensive experiments, benchmarking our method against cutting-edge models such as Hunyuan-Avatar and Omnihuman. The experimental results consistently demonstrate that our approach significantly outperforms these existing solutions. Additionally, we explore the versatility of our method through its applications in long-form video generation and precise video lip-sync editing.

cs.CV↗

arXiv2025

Harnessing Large Language Models for Disaster Management: A Survey

Large language models (LLMs) have revolutionized scientific research with their exceptional capabilities and transformed various fields. Among their practical applications, LLMs have been playing a crucial role in mitigating threats to human life, infrastructure, and the environment. Despite growing research in disaster LLMs, there remains a lack of systematic review and in-depth analysis of LLMs for natural disaster management. To address the gap, this paper presents a comprehensive survey of existing LLMs in natural disaster management, along with a taxonomy that categorizes existing works based on disaster phases and application scenarios. By collecting public datasets and identifying key challenges and opportunities, this study aims to guide the professional community in developing advanced LLMs for disaster management to enhance the resilience against natural disasters.

cs.CL↗

arXiv2025

Novel discretization method to calculate g-functions of vertical geothermal boreholes with improved accuracy and efficiency

The calculation of g-functions is essential for the design and simulation of geothermal boreholes. However, existing methods, such as the stacked finite line source (SFLS) model, face challenges regarding computational efficiency and accuracy, particularly with fine-grained discretization. This paper introduces a novel discretization method to address these limitations. We reformulate the g-function calculation under the uniform borehole wall temperature boundary condition as the solution to spatio-temporal integral equations. The SFLS model is identified as a special case using stepwise approximation of the heat extraction rate. Our proposed method employs the Gauss-Legendre quadrature to approximate the spatial integrals with a weighted sum of function values at strategically chosen points. This transforms the time-consuming segment-to-segment integral calculations in SFLS model into simpler and analytical point-to-point response factors. Furthermore, we identify that the governing integral equations are of the Fredholm first kind, leading to ill-conditioned linear systems that can cause g-function to diverge at high discretization orders. To address this, a regularization technique is implemented to ensure stable and convergent solutions. Numerical tests demonstrate that the proposed method is significantly more efficient, achieving comparable or improved accuracy at speeds 20 to 200 times faster than the SFLS model with optimized nonuniform discretization schemes.

physics.comp-ph↗

arXiv2025

Scaling behaviour of rotating convection in a spherical shell with different Prandtl numbers

Rayleigh-Benard convection in a rotating spherical shell provides a simplified model for convective dynamics of planetary and stellar interiors. In this study, we build more than 200 numerical models of rotating convection in a spherical shell over a wide range of $\Pran$ ($10^{-2}\le\Pran\le10^2$). As increasing the Rayleigh number $Ra$, we characterise four different flow regimes, starting from the linear onset to multiple modes, then transiting to the geostrophic turbulence and eventually approaching the weakly rotating regime. In the multiple modes regime, we show evidence of triadic resonances in numerical models with different $\Pran$, which may provide a generic mechanism for the transition from laminar to turbulence in rotating convection. We analyse scaling behaviours of the heat transfer and convective flow speeds in numerical simulations, paying particular attention to the $\Pran$-dependence. We find that the so-called diffusion-free scaling for the heat transfer cannot reconcile all numerical models with different $\Pran$ in the geostrophic turbulence regime. However, the characteristic flow speeds at different $\Pran$ roughly follow a unified scaling that can be described by VAC force balances, though the scaling tends to approach the CIA force balance at low $\Pran$. Both scaling behaviours and transition behaviours suggest that the heat transfer is controlled by the boundary layers while the convective flow speeds are mainly determined by the force balance in the bulk for cases with $\Pran>1$, which is in line with recent experimental results with moderate to high $\Pran$. For cases with $\Pran \le 1$, both the heat transfer and convective velocities are approaching the inviscid dynamics in the bulk.

physics.flu-dyn↗

arXiv2025

Onsager Principle-Based Domain Embedding for Thermodynamically Consistent Cahn-Hilliard Model in Arbitrary Domain

The original Cahn-Hilliard model in an arbitrary domain with two prescribed boundary conditions is extended to a Cahn-Hilliard-type model in a larger, regular domain with homogeneous Neumann boundary conditions. The extension is based on the Onsager principle-based domain embedding (OPBDE) method, which has been developed as a systematic domain embedding framework to ensure thermodynamic consistency. By introducing a modified conservation law, the flux at the boundary of the original domain is incorporated into the conservation law as a source term. Our variational approach demonstrates that, even without a prior knowledge on the specific form of the rate of free energy pumped into the system, the Onsager principle remains an effective instrument in deriving the constitutive equation of the extended system. This approach clarifies the intrinsic structure of the extended model in the perspectives of free energy and its dissipation. Asymptotic analysis is carried out for the extended OPBDE Cahn-Hilliard model, demonstrating that the original Cahn-Hilliard model, including its boundary conditions, can be fully recovered. To validate our approach, a structure-preserving numerical scheme is developed to discretize the extended model. Numerical results show that the OPBDE Cahn-Hilliard model is accurate, effective, and robust, highlighting the capability of the OPBDE method in handling gradient flow problems in arbitrary domain geometries.

math.NA↗

arXiv2025

AR-GRPO: Training Autoregressive Image Generation Models via Reinforcement Learning

Inspired by the success of reinforcement learning (RL) in refining large language models (LLMs), we propose AR-GRPO, an approach to integrate online RL training into autoregressive (AR) image generation models. We adapt the Group Relative Policy Optimization (GRPO) algorithm to refine the vanilla autoregressive models' outputs by carefully designed reward functions that evaluate generated images across multiple quality dimensions, including perceptual quality, realism, and semantic fidelity. We conduct comprehensive experiments on both class-conditional (i.e., class-to-image) and text-conditional (i.e., text-to-image) image generation tasks, demonstrating that our RL-enhanced framework significantly improves both the image quality and human preference of generated images compared to the standard AR baselines. Our results show consistent improvements across various evaluation metrics, establishing the viability of RL-based optimization for AR image generation and opening new avenues for controllable and high-quality image synthesis. The source codes and models are available at: https://github.com/Kwai-Klear/AR-GRPO.

cs.CV↗

arXiv2025

TopKD: Top-scaled Knowledge Distillation

Recent advances in knowledge distillation (KD) predominantly emphasize feature-level knowledge transfer, frequently overlooking critical information embedded within the teacher's logit distributions. In this paper, we revisit logit-based distillation and reveal an underexplored yet critical element: Top-K knowledge. Motivated by this insight, we propose Top-scaled Knowledge Distillation (TopKD), a simple, efficient, and architecture-agnostic framework that significantly enhances logit-based distillation. TopKD consists of two main components: (1) a Top-K Scaling Module (TSM), which adaptively amplifies the most informative logits, and (2) a Top-K Decoupled Loss (TDL), which offers targeted and effective supervision. Notably, TopKD integrates seamlessly into existing KD methods without introducing extra modules or requiring architectural changes. Extensive experiments on CIFAR-100, ImageNet, STL-10, and Tiny-ImageNet demonstrate that TopKD consistently surpasses state-of-the-art distillation methods. Moreover, our method demonstrates substantial effectiveness when distilling Vision Transformers, underscoring its versatility across diverse network architectures. These findings highlight the significant potential of logits to advance knowledge distillation.

cs.CV↗