arXiv subjects

Yue Wang

Publications and source records attributed to Yue Wang.

At least 325 records · Page 18Linked to original sources

arXiv2025

Capsizing-Guided Trajectory Optimization for Autonomous Navigation with Rough Terrain

It is a challenging task for ground robots to autonomously navigate in harsh environments due to the presence of non-trivial obstacles and uneven terrain. This requires trajectory planning that balances safety and efficiency. The primary challenge is to generate a feasible trajectory that prevents robot from tip-over while ensuring effective navigation. In this paper, we propose a capsizing-aware trajectory planner (CAP) to achieve trajectory planning on the uneven terrain. The tip-over stability of the robot on rough terrain is analyzed. Based on the tip-over stability, we define the traversable orientation, which indicates the safe range of robot orientations. This orientation is then incorporated into a capsizing-safety constraint for trajectory optimization. We employ a graph-based solver to compute a robust and feasible trajectory while adhering to the capsizing-safety constraint. Extensive simulation and real-world experiments validate the effectiveness and robustness of the proposed method. The results demonstrate that CAP outperforms existing state-of-the-art approaches, providing enhanced navigation performance on uneven terrains.

cs.RO↗

arXiv2025

Learning an Implicit Physics Model for Image-based Fluid Simulation

Humans possess an exceptional ability to imagine 4D scenes, encompassing both motion and 3D geometry, from a single still image. This ability is rooted in our accumulated observations of similar scenes and an intuitive understanding of physics. In this paper, we aim to replicate this capacity in neural networks, specifically focusing on natural fluid imagery. Existing methods for this task typically employ simplistic 2D motion estimators to animate the image, leading to motion predictions that often defy physical principles, resulting in unrealistic animations. Our approach introduces a novel method for generating 4D scenes with physics-consistent animation from a single image. We propose the use of a physics-informed neural network that predicts motion for each surface point, guided by a loss term derived from fundamental physical principles, including the Navier-Stokes equations. To capture appearance, we predict feature-based 3D Gaussians from the input image and its estimated depth, which are then animated using the predicted motions and rendered from any desired camera perspective. Experimental results highlight the effectiveness of our method in producing physically plausible animations, showcasing significant performance improvements over existing methods. Our project page is https://physfluid.github.io/ .

cs.CV↗

arXiv2025

Braided mixing in confined chiral active matter

Efficient mixing of fluids is essential in many practical applications to achieve homogeneity. For microscopic systems, however, both diffusion and turbulence are ineffective methods to achieve chaotic mixing due to the low Reynolds number, hence either active stirring or inducing turbulence through geometric boundary effects are generally implemented. Here, we study a modified chiral Vicsek model, where active microswimmers act as moving rods, stirring the surrounding substrate. We study the degree of mixing in the patterns formed by interplay between confinement, chiral motion and alignment interactions. This mixing is computed by considering the entanglement of spacetime trajectories of the particles, which forms a braid. Optimising the finite-time braiding exponent of this braid then yields a set of constituent parameters of the system, showing that a pattern consisting of a local stable vortex droplet and an ordered oscillating phase achieves the highest degree of mixing.

cond-mat.soft↗

arXiv2025

A Whole-Body Motion Imitation Framework from Human Data for Full-Size Humanoid Robot

Motion imitation is a pivotal and effective approach for humanoid robots to achieve a more diverse range of complex and expressive movements, making their performances more human-like. However, the significant differences in kinematics and dynamics between humanoid robots and humans present a major challenge in accurately imitating motion while maintaining balance. In this paper, we propose a novel whole-body motion imitation framework for a full-size humanoid robot. The proposed method employs contact-aware whole-body motion retargeting to mimic human motion and provide initial values for reference trajectories, and the non-linear centroidal model predictive controller ensures the motion accuracy while maintaining balance and overcoming external disturbances in real time. The assistance of the whole-body controller allows for more precise torque control. Experiments have been conducted to imitate a variety of human motions both in simulation and in a real-world humanoid robot. These experiments demonstrate the capability of performing with accuracy and adaptability, which validates the effectiveness of our approach.

cs.RO↗

arXiv2025

DiSC-Med: Diffusion-based Semantic Communications for Robust Medical Image Transmission

The rapid development of artificial intelligence has driven smart health with next-generation wireless communication technologies, stimulating exciting applications in remote diagnosis and intervention. To enable a timely and effective response for remote healthcare, efficient transmission of medical data through noisy channels with limited bandwidth emerges as a critical challenge. In this work, we propose a novel diffusion-based semantic communication framework, namely DiSC-Med, for the medical image transmission, where medical-enhanced compression and denoising blocks are developed for bandwidth efficiency and robustness, respectively. Unlike conventional pixel-wise communication framework, our proposed DiSC-Med is able to capture the key semantic information and achieve superior reconstruction performance with ultra-high bandwidth efficiency against noisy channels. Extensive experiments on real-world medical datasets validate the effectiveness of our framework, demonstrating its potential for robust and efficient telehealth applications.

cs.LG↗

arXiv2025

Mmm whatcha say? Uncovering distal and proximal context effects in first and second-language word perception using psychophysical reverse correlation

Acoustic context effects, where surrounding changes in pitch, rate or timbre influence the perception of a sound, are well documented in speech perception, but how they interact with language background remains unclear. Using a reverse-correlation approach, we systematically varied the pitch and speech rate in phrases around different pairs of vowels for second language (L2) speakers of English (/i/-/I/) and French (/u/-/y/), thus reconstructing, in a data-driven manner, the prosodic profiles that bias their perception. Testing English and French speakers (n=25), we showed that vowel perception is in fact influenced by conflicting effects from the surrounding pitch and speech rate: a congruent proximal effect 0.2s pre-target and a distal contrastive effect up to 1s before; and found that L1 and L2 speakers exhibited strikingly similar prosodic profiles in perception. We provide a novel method to investigate acoustic context effects across stimuli, timescales, and acoustic domain.

cs.SD↗

arXiv2025

Advancing Multimodal Reasoning via Reinforcement Learning with Cold Start

Recent advancements in large language models (LLMs) have demonstrated impressive chain-of-thought reasoning capabilities, with reinforcement learning (RL) playing a crucial role in this progress. While "aha moment" patterns--where models exhibit self-correction through reflection--are often attributed to emergent properties from RL, we first demonstrate that these patterns exist in multimodal LLMs (MLLMs) prior to RL training but may not necessarily correlate with improved reasoning performance. Building on these insights, we present a comprehensive study on enhancing multimodal reasoning through a two-stage approach: (1) supervised fine-tuning (SFT) as a cold start with structured chain-of-thought reasoning patterns, followed by (2) reinforcement learning via GRPO to further refine these capabilities. Our extensive experiments show that this combined approach consistently outperforms both SFT-only and RL-only methods across challenging multimodal reasoning benchmarks. The resulting models achieve state-of-the-art performance among open-source MLLMs at both 3B and 7B scales, with our 7B model showing substantial improvements over base models (e.g., 66.3 %$\rightarrow$73.4 % on MathVista, 62.9 %$\rightarrow$70.4 % on We-Math) and our 3B model achieving performance competitive with several 7B models. Overall, this work provides practical guidance for building advanced multimodal reasoning models. Our code is available at https://github.com/waltonfuture/RL-with-Cold-Start.

cs.CL↗

arXiv2025

Generalized Advantage Estimation for Distributional Policy Gradients

Generalized Advantage Estimation (GAE) has been used to mitigate the computational complexity of reinforcement learning (RL) by employing an exponentially weighted estimation of the advantage function to reduce the variance in policy gradient estimates. Despite its effectiveness, GAE is not designed to handle value distributions integral to distributional RL, which can capture the inherent stochasticity in systems and is hence more robust to system noises. To address this gap, we propose a novel approach that utilizes the optimal transport theory to introduce a Wasserstein-like directional metric, which measures both the distance and the directional discrepancies between probability distributions. Using the exponentially weighted estimation, we leverage this Wasserstein-like directional metric to derive distributional GAE (DGAE). Similar to traditional GAE, our proposed DGAE provides a low-variance advantage estimate with controlled bias, making it well-suited for policy gradient algorithms that rely on advantage estimation for policy updates. We integrated DGAE into three different policy gradient methods. Algorithms were evaluated across various OpenAI Gym environments and compared with the baselines with traditional GAE to assess the performance.

cs.LG↗

arXiv2025

Chemical Treatment-Induced Indirect-to-Direct Bandgap Transition in MoS2: Impact on Optical Properties

The unique electrical and optical properties of emerging two-dimensional transition metal dichal-cogenides (TMDs) present compelling advantages over conventional semiconductors, including Si, Ge, and GaAs. Nevertheless, realising the full potential of TMDs in electronic and optoelectronic devices, such as transistors, light-emitting diodes (LEDs), and photodetectors, is con-strained by high contact resistance. This limitation arises from their low intrinsic carrier concen-trations and the current insufficiency of doping strategies for atomically thin materials. Notably, chemical treatment with 1,2-dichloroethane (DCE) has been demonstrated as an effective post-growth method to enhance the n-type electrical conductivity of TMDs. Despite the well-documented electrical improvements post-DCE treatment, its effects on optical properties, specifically the retention of optical characteristics and excitonic behaviour, are not yet clearly under-stood. Here, we systematically investigate the layer- and time-dependent optical effects of DCE on molybdenum disulfide (MoS2) using photoluminescence (PL) spectroscopy and Density Functional Theory (DFT) simulations. Our PL results reveal a rapid reduction in the indirect bandgap transition, with the direct transition remaining unaffected. DFT confirms that chlorine (Cl) atoms bind to sulphur vacancies, creating mid-gap states that facilitate non-radiative recom-bination, explaining the observed indirect PL suppression. This work demonstrates DCE's utility not only for n-type doping but also for optical band structure engineering in MoS2 by selec-tively suppressing indirect transitions, potentially opening new avenues for 2D optoelectronic device design.

cond-mat.mtrl-sci↗

arXiv2025

Aria-UI: Visual Grounding for GUI Instructions

Digital agents for automating tasks across different platforms by directly manipulating the GUIs are increasingly important. For these agents, grounding from language instructions to target elements remains a significant challenge due to reliance on HTML or AXTree inputs. In this paper, we introduce Aria-UI, a large multimodal model specifically designed for GUI grounding. Aria-UI adopts a pure-vision approach, eschewing reliance on auxiliary inputs. To adapt to heterogeneous planning instructions, we propose a scalable data pipeline that synthesizes diverse and high-quality instruction samples for grounding. To handle dynamic contexts in task performing, Aria-UI incorporates textual and text-image interleaved action histories, enabling robust context-aware reasoning for grounding. Aria-UI sets new state-of-the-art results across offline and online agent benchmarks, outperforming both vision-only and AXTree-reliant baselines. We release all training data and model checkpoints to foster further research at https://ariaui.github.io.

cs.HC↗

arXiv2025

Chat2SPaT: A Large Language Model Based Tool for Automating Traffic Signal Control Plan Management

Pre-timed traffic signal control, commonly used for operating signalized intersections and coordinated arterials, requires tedious manual work for signaling plan creating and updating. When the time-of-day or day-of-week plans are utilized, one intersection is often associated with multiple plans, leading to further repetitive manual plan parameter inputting. To enable a user-friendly traffic signal control plan management process, this study proposes Chat2SPaT, a method to convert users' semi-structured and ambiguous descriptions on the signal control plan to exact signal phase and timing (SPaT) results, which could further be transformed into structured stage-based or ring-based plans to interact with intelligent transportation system (ITS) software and traffic signal controllers. With curated prompts, Chat2SPaT first leverages large language models' (LLMs) capability of understanding users' plan descriptions and reformulate the plan as a combination of phase sequence and phase attribute results in the json format. Based on LLM outputs, python scripts are designed to locate phases in a cycle, address nuances of traffic signal control, and finally assemble the complete traffic signal control plan. Within a chat, the pipeline can be utilized iteratively to conduct further plan editing. Experiments show that Chat2SPaT can generate plans with an accuracy of over 94% for both English and Chinese cases, using a test dataset with over 300 plan descriptions. As the first benchmark for evaluating LLMs' capability of understanding traffic signal control plan descriptions, Chat2SPaT provides an easy-to-use plan management pipeline for traffic practitioners and researchers, serving as a potential new building block for a more accurate and versatile application of LLMs in the field of ITS. The source codes, prompts and test dataset are openly accessible at https://github.com/yuewangits/Chat2SPaT.

cs.AI↗

arXiv2025

InstantSplat: Sparse-view Gaussian Splatting in Seconds

While neural 3D reconstruction has advanced substantially, its performance significantly degrades with sparse-view data, which limits its broader applicability, since SfM is often unreliable in sparse-view scenarios where feature matches are scarce. In this paper, we introduce InstantSplat, a novel approach for addressing sparse-view 3D scene reconstruction at lightning-fast speed. InstantSplat employs a self-supervised framework that optimizes 3D scene representation and camera poses by unprojecting 2D pixels into 3D space and aligning them using differentiable neural rendering. The optimization process is initialized with a large-scale trained geometric foundation model, which provides dense priors that yield initial points through model inference, after which we further optimize all scene parameters using photometric errors. To mitigate redundancy introduced by the prior model, we propose a co-visibility-based geometry initialization, and a Gaussian-based bundle adjustment is employed to rapidly adapt both the scene representation and camera parameters without relying on a complex adaptive density control process. Overall, InstantSplat is compatible with multiple point-based representations for view synthesis and surface reconstruction. It achieves an acceleration of over 30x in reconstruction and improves visual quality (SSIM) from 0.3755 to 0.7624 compared to traditional SfM with 3D-GS.

cs.CV↗

arXiv2025

CoMMiT: Co-informed inference of microbiome-metabolome interactions via transfer learning

Recent multi-omic microbiome studies enable integrative analysis of microbes and metabolites, uncovering their associations with various host conditions. Such analyses require multivariate models capable of accounting for the complex correlation structures between microbes and metabolites. However, existing multivariate models often suffer from low statistical power for detecting microbiome-metabolome interactions due to small sample sizes and weak biological signals. To address these challenges, we introduce CoMMiT, Co-informed inference of Microbiome-Metabolome Interactions via novel Transfer learning models. Unlike conventional transfer-learning methods that borrow information from external datasets, CoMMiT leverages similarities across metabolites within a single cohort, reducing the risk of negative transfer often caused by differences in sequencing platforms and bioinformatic pipelines across studies. CoMMiT operates under the flexible assumption that auxiliary metabolites are collectively informative for the target metabolite, without requiring individual auxiliary metabolites to be informative. CoMMiT uses a novel data-driven approach to selecting the optimal set of auxiliary metabolites. Using this optimal set, CoMMiT employs a de-biasing framework to enable efficient calculation of p-values, facilitating the identification of statistically significant microbiome-metabolome interactions. Applying CoMMiT to a feeding study reveals biologically meaningful microbiome-metabolome interactions under a low glycemic load diet, demonstrating the diet-host link through gut metabolism.

stat.ME↗

arXiv2025

InfiniCube: Unbounded and Controllable Dynamic 3D Driving Scene Generation with World-Guided Video Models

We present InfiniCube, a scalable method for generating unbounded dynamic 3D driving scenes with high fidelity and controllability. Previous methods for scene generation either suffer from limited scales or lack geometric and appearance consistency along generated sequences. In contrast, we leverage the recent advancements in scalable 3D representation and video models to achieve large dynamic scene generation that allows flexible controls through HD maps, vehicle bounding boxes, and text descriptions. First, we construct a map-conditioned sparse-voxel-based 3D generative model to unleash its power for unbounded voxel world generation. Then, we re-purpose a video model and ground it on the voxel world through a set of carefully designed pixel-aligned guidance buffers, synthesizing a consistent appearance. Finally, we propose a fast feed-forward approach that employs both voxel and pixel branches to lift the dynamic videos to dynamic 3D Gaussians with controllable objects. Our method can generate controllable and realistic 3D driving scenes, and extensive experiments validate the effectiveness and superiority of our model.

cs.CV↗

arXiv2025

Wavelet Diffusion Neural Operator

Simulating and controlling physical systems described by partial differential equations (PDEs) are crucial tasks across science and engineering. Recently, diffusion generative models have emerged as a competitive class of methods for these tasks due to their ability to capture long-term dependencies and model high-dimensional states. However, diffusion models typically struggle with handling system states with abrupt changes and generalizing to higher resolutions. In this work, we propose Wavelet Diffusion Neural Operator (WDNO), a novel PDE simulation and control framework that enhances the handling of these complexities. WDNO comprises two key innovations. Firstly, WDNO performs diffusion-based generative modeling in the wavelet domain for the entire trajectory to handle abrupt changes and long-term dependencies effectively. Secondly, to address the issue of poor generalization across different resolutions, which is one of the fundamental tasks in modeling physical systems, we introduce multi-resolution training. We validate WDNO on five physical systems, including 1D advection equation, three challenging physical systems with abrupt changes (1D Burgers' equation, 1D compressible Navier-Stokes equation and 2D incompressible fluid), and a real-world dataset ERA5, which demonstrates superior performance on both simulation and control tasks over state-of-the-art methods, with significant improvements in long-term and detail prediction accuracy. Remarkably, in the challenging context of the 2D high-dimensional and indirect control task aimed at reducing smoke leakage, WDNO reduces the leakage by 78% compared to the second-best baseline. The code can be found at https://github.com/AI4Science-WestlakeU/wdno.git.

cs.LG↗

arXiv2025

Real-time Terrain Analysis for Off-road Autonomous Vehicles

This research addresses critical autonomous vehicle control challenges arising from road roughness variation, which induces course deviations and potential loss of road contact during steering operations. We present a novel real-time road roughness estimation system employing Bayesian calibration methodology that processes axle accelerations to predict terrain roughness with quantifiable confidence measures. The technical framework integrates a Gaussian process surrogate model with a simulated half-vehicle model, systematically processing vehicle velocity and road surface roughness parameters to generate corresponding axle acceleration responses. The Bayesian calibration routine performs inverse estimation of road roughness from observed accelerations and velocities, yielding posterior distributions that quantify prediction uncertainty for adaptive risk management. Training data generation utilizes Latin Hypercube sampling across comprehensive velocity and roughness parameter spaces, while the calibrated model integrates seamlessly with a Simplex controller architecture to dynamically adjust velocity limits based on real-time roughness predictions. Experimental validation on stochastically generated surfaces featuring varying roughness regions demonstrates robust real-time characterization capabilities, with the integrated Simplex control strategy effectively enhancing autonomous vehicle operational safety through proactive surface condition response. This innovative Bayesian framework establishes a comprehensive foundation for mitigating roughness-related operational risks while simultaneously improving efficiency and safety margins in autonomous vehicle systems.

cs.RO↗

arXiv2025

No Free Lunch: Rethinking Internal Feedback for LLM Reasoning

Reinforcement learning has emerged as a powerful paradigm for post-training large language models (LLMs) to improve reasoning. Approaches like Reinforcement Learning from Human Feedback (RLHF) and Reinforcement Learning with Verifiable Rewards (RLVR) have shown strong results, but they require extensive external supervision. We investigate an alternative class of methods, Reinforcement Learning from Internal Feedback (RLIF), which relies solely on intrinsic model-derived signals instead of external rewards. In particular, we leverage unsupervised reward proxies such as token-level entropy, trajectory-level entropy, and self-certainty. Our theoretical analysis shows these internal objectives are partially equivalent, and we empirically evaluate various RLIF strategies on challenging math reasoning benchmarks. Experimental results demonstrate that RLIF can boost the reasoning performance of base LLMs at the beginning phase of the training, matching or surpassing RLVR techniques on these tasks. However, when training progresses, performance degrades even below the model before training. Moreover, we find that RLIF yields little improvement for instruction-tuned models, indicating diminishing returns of intrinsic feedback once an LLM is already instruction-tuned. We further analyze this limitation by mixing model weights and explain the reason of RLIF's training behaviors, providing practical guidelines for integrating internal feedback signals into LLM training. We hope our analysis of internal feedback will inform more principled and effective strategies for LLM post-training.

cs.LG↗

arXiv2025

SEED: A Structural Encoder for Embedding-Driven Decoding in Time Series Prediction with LLMs

Multivariate time series forecasting requires models to simultaneously capture variable-wise structural dependencies and generalize across diverse tasks. While structural encoders are effective in modeling feature interactions, they lack the capacity to support semantic-level reasoning or task adaptation. Conversely, large language models (LLMs) possess strong generalization capabilities but remain incompatible with raw time series inputs. This gap limits the development of unified, transferable prediction systems. Therefore, we introduce SEED, a structural encoder for embedding-driven decoding, which integrates four stages: a token-aware encoder for patch extraction, a projection module that aligns patches with language model embeddings, a semantic reprogramming mechanism that maps patches to task-aware prototypes, and a frozen language model for prediction. This modular architecture decouples representation learning from inference, enabling efficient alignment between numerical patterns and semantic reasoning. Empirical results demonstrate that the proposed method achieves consistent improvements over strong baselines, and comparative studies on various datasets confirm SEED's role in addressing the structural-semantic modeling gap.

cs.CL↗