arXiv subjects

Zhicheng Wang

Publications and source records attributed to Zhicheng Wang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Explore More, Learn Better: Parallel MLLM Embeddings under Mutual Information Minimization

Embedding models are a cornerstone of modern AI. Driven by Multimodal Large Language Models (MLLMs), they have made great progress in architecture and data curation, while the holistic paradigm is still limited to SSC, i.e., single input, singular embedding, contrastive supervision, which collapses rich, multifaceted inputs into monolithic embeddings and fails to fully exploit MLLM capabilities. In this paper, we tailor one Parallel Decoupling Framework (PDF) for multimodal embedding learning, by utilizing the proprietary steerability of MLLMs, i.e., their ability to flexibly generate quite differentiated response under explicit instructions. Concretely, PDF conditions a shared MLLM backbone on distinct, learnable prefixes to roll out multiple parallel paths for one input, then relies on these paths to obtain parallel embeddings. To promote full parallel diversity, we employ Mutual Information Minimization (MIM) as an explicit constraint, coupled with per-path contrastive supervision to maintain semantic alignment. Such dual-objectives force PDF to yield robust semantic coverage and a generalizable embedding space. Ultimately, the remarkable embedding space are accessible at inference via one single forward pass, incurring negligible computational overhead. We instantiate PDF on multiple MLLM backbones and prove its effectiveness on MMEB benchmark. Significant gains are consistently achieved across various resolutions and model sizes, e.g., boosting the VLM2Vec-LLaVA-1.6-LR model by a remarkable +8.9% (7B), while the VLM2Vec-Qwen2VL models by +4.2% (2B) and +3.1% (7B). In terms of efficiency, our 2B model surpasses its baseline by +2.6% using only half the computational budget.

cs.LG↗

arXiv2025

Robot Learning from a Physical World Model

We introduce PhysWorld, a framework that enables robot learning from video generation through physical world modeling. Recent video generation models can synthesize photorealistic visual demonstrations from language commands and images, offering a powerful yet underexplored source of training signals for robotics. However, directly retargeting pixel motions from generated videos to robots neglects physics, often resulting in inaccurate manipulations. PhysWorld addresses this limitation by coupling video generation with physical world reconstruction. Given a single image and a task command, our method generates task-conditioned videos and reconstructs the underlying physical world from the videos, and the generated video motions are grounded into physically accurate actions through object-centric residual reinforcement learning with the physical world model. This synergy transforms implicit visual guidance into physically executable robotic trajectories, eliminating the need for real robot data collection and enabling zero-shot generalizable robotic manipulation. Experiments on diverse real-world tasks demonstrate that PhysWorld substantially improves manipulation accuracy compared to previous approaches. Visit \href{https://pointscoder.github.io/PhysWorld_Web/}{the project webpage} for details.

cs.RO↗

arXiv2025

Fast Measuring Pavement Crack Width by Cascading Principal Component Analysis

Accurate quantification of pavement crack width plays a pivotal role in assessing structural integrity and guiding maintenance interventions. However, achieving precise crack width measurements presents significant challenges due to: (1) the complex, non-uniform morphology of crack boundaries, which limits the efficacy of conventional approaches, and (2) the demand for rapid measurement capabilities from arbitrary pixel locations to facilitate comprehensive pavement condition evaluation. To overcome these limitations, this study introduces a cascaded framework integrating Principal Component Analysis (PCA) and Robust PCA (RPCA) for efficient crack width extraction from digital images. The proposed methodology comprises three sequential stages: (1) initial crack segmentation using established detection algorithms to generate a binary representation, (2) determination of the primary orientation axis for quasi-parallel cracks through PCA, and (3) extraction of the Main Propagation Axis (MPA) for irregular crack geometries using RPCA. Comprehensive evaluations were conducted across three publicly available datasets, demonstrating that the proposed approach achieves superior performance in both computational efficiency and measurement accuracy compared to existing state-of-the-art techniques.

cs.CV↗

arXiv2025

Wave-Particle (Continuous-Discrete) Dualistic Visual Tokenization for Unified Understanding and Generation

The unification of understanding and generation within a single multi-modal large model (MLLM) remains one significant challenge, largely due to the dichotomy between continuous and discrete visual tokenizations. Continuous tokenizer (CT) achieves strong performance by bridging multiple independently-trained understanding modules and generation modules, but suffers from complex multi-stage pipelines and substantial engineering overhead. Conversely, discrete tokenizers (DT) offer a conceptually elegant idea by quantizing each image into a primitive, but inevitably leading to information loss and performance degradation. To resolve this tension, we question the binary choice between CT and DT, inspired by the wave-particle duality of light, and propose the Continuous-Discrete Dualistic Visual Tokenizer (CDD-VT). We treat visual data as a flexible composition of image primitives derived from quantized codebooks, with the crucial insight that the primitive number assigned to each visual sample is adaptively determined according to its complexity: simple instances use a few primitives, emulating discrete tokenization, while complex instances use many, approximating continuous tokenization. Two core components are designed: Diverse Quantitative Primitives, which encourage primitives orthogonality to better populate information space, and Dynamic Primitive Allocator, which assesses sample complexity to determine the optimal set of primitives. Extensive experiments on reconstruction, retrieval and classification show that CDD-VT achieves superior performance over to specialized CT and DT, effectively getting strong result within a concise and scalable MLLM.

cs.CV↗

arXiv2025

Rectify and Align GPS Points to Parking Spots via Rank-1 Constraint

Parking spots are essential components, providing vital mobile resources for residents in a city. Accurate Global Positioning System (GPS) points of parking spots are the core data for subsequent applications,e.g., parking management, parking policy, and urban development. However, high-rise buildings tend to cause GPS points to drift from the actual locations of parking spots; besides, the standard lower-cost GPS equipment itself has a certain location error. Therefore, it is a non-trivial task to correct a few wrong GPS points from a large number of parking spots in an unsupervised approach. In this paper, motivated by the physical constraints of parking spots (i.e., parking spots are parallel to the sides of roads), we propose an unsupervised low-rank method to effectively rectify errors in GPS points and further align them to the parking spots in a unified framework. The proposed unconventional rectification and alignment method is simple and yet effective for any type of GPS point errors. Extensive experiments demonstrate the superiority of the proposed method to solve a practical problem. The data set and the code are publicly accessible at:https://github.com/pangjunbiao/ITS-Parking-spots-Dataset.

cs.LG↗

arXiv2025

GaussianLens: Localized High-Resolution Reconstruction via On-Demand Gaussian Densification

We perceive our surroundings with an active focus, paying more attention to regions of interest, such as the shelf labels in a grocery store. When it comes to scene reconstruction, this human perception trait calls for spatially varying degrees of detail ready for closer inspection in critical regions, preferably reconstructed on demand. While recent works in 3D Gaussian Splatting (3DGS) achieve fast, generalizable reconstruction from sparse views, their uniform resolution output leads to high computational costs unscalable to high-resolution training. As a result, they cannot leverage available images at their original high resolution to reconstruct details. Per-scene optimization methods reconstruct finer details with adaptive density control, yet require dense observations and lengthy offline optimization. To bridge the gap between the prohibitive cost of high-resolution holistic reconstructions and the user needs for localized fine details, we propose the problem of localized high-resolution reconstruction via on-demand Gaussian densification. Given a low-resolution 3DGS reconstruction, the goal is to learn a generalizable network that densifies the initial 3DGS to capture fine details in a user-specified local region of interest (RoI), based on sparse high-resolution observations of the RoI. This formulation avoids the high cost and redundancy of uniformly high-resolution reconstructions and fully leverages high-resolution captures in critical regions. We propose GaussianLens, a feed-forward densification framework that fuses multi-modal information from the initial 3DGS and multi-view images. We further design a pixel-guided densification mechanism that effectively captures details under large resolution increases. Experiments demonstrate our method's superior performance in local fine detail reconstruction and strong scalability to images of up to $1024\times1024$ resolution.

cs.CV↗

arXiv2025

Decoding in Latent Spaces for Efficient Inference in LLM-based Recommendation

Fine-tuning large language models (LLMs) for recommendation in a generative manner has delivered promising results, but encounters significant inference overhead due to autoregressive decoding in the language space. This work explores bypassing language-space decoding by directly matching candidate items with the LLM's internal thought representations in the latent space, eliminating the time-consuming autoregressive process to reduce computational costs. Towards this, we introduce Light Latent-space Decoding (L2D), an effective and efficient latent-space decoding method. L2D represents user-preferred items by using the hidden states of test sequences reflecting the LLM's internal thought, and obtains candidate item representations from the hidden states of training sequences labeled with the corresponding candidate items. It then matches the two types of representations to decode items, achieving latent-space decoding. In this way, it enables efficient decoding without altering the LLM's generative tuning paradigm, thereby preserving performance. Extensive empirical results demonstrate that L2D is more than 10x faster than language-space decoding while maintaining or enhancing performance.

cs.IR↗

arXiv2025

Learning Turbulent Flows with Generative Models: Super-resolution, Forecasting, and Sparse Flow Reconstruction

Neural operators are promising surrogates for dynamical systems but when trained with standard L2 losses they tend to oversmooth fine-scale turbulent structures. Here, we show that combining operator learning with generative modeling overcomes this limitation. We consider three practical turbulent-flow challenges where conventional neural operators fail: spatio-temporal super-resolution, forecasting, and sparse flow reconstruction. For Schlieren jet super-resolution, an adversarially trained neural operator (adv-NO) reduces the energy-spectrum error by 15x while preserving sharp gradients at neural operator-like inference cost. For 3D homogeneous isotropic turbulence, adv-NO trained on only 160 timesteps from a single trajectory forecasts accurately for five eddy-turnover times and offers 114x wall-clock speed-up at inference than the baseline diffusion-based forecasters, enabling near-real-time rollouts. For reconstructing cylinder wake flows from highly sparse Particle Tracking Velocimetry-like inputs, a conditional generative model infers full 3D velocity and pressure fields with correct phase alignment and statistics. These advances enable accurate reconstruction and forecasting at low compute cost, bringing near-real-time analysis and control within reach in experimental and computational fluid mechanics. See our project page: https://vivekoommen.github.io/Gen4Turb/

physics.flu-dyn↗

arXiv2025

Versatile silicon integrated photonic processor: a reconfigurable solution for next-generation AI clusters

The Artificial Intelligence models pose serious challenges in intensive computing and high-bandwidth communication for conventional electronic circuit-based computing clusters. Silicon photonic technologies, owing to their high speed, low latency, large bandwidth, and complementary metal-oxide-semiconductor compatibility, have been widely implemented for data transfer and actively explored as photonic neural networks in AI clusters. However, current silicon photonic integrated chips lack adaptability for multifuncional use and hardware-software systematic coordination. Here, we develop a reconfigurable silicon photonic processor with $40$ programmable unit cells integrating over $160$ component, which, to the best of our knowledge, is the first to realize diverse functions with a chip for AI clusters, from computing acceleration and signal processing to network swtiching and secure encryption. Through a self-developed automated testing, compilation, and tuning framework to the processor without in-network monitoring photodetectors, we implement $4\times4$ dual-direction unitary and $3\times3$ uni-direction non-unitary matrix multiplications, neural networks for image recognition, micro-ring modulator wavelength locking, $4\times4$ photonic channel switching , and silicon photonic physical unclonable functions. This optoelectronic processing system, incorporating the photonic processor and its software stack, paves the way for both advanced photonic system-on-chip design and the construction of photo-electronic AI clusters.

physics.optics↗

arXiv2025

ALow-Cost Real-Time Framework for Industrial Action Recognition Using Foundation Models

Action recognition (AR) in industrial environments -- particularly for identifying actions and operational gestures -- faces persistent challenges due to high deployment costs, poor cross-scenario generalization, and limited real-time performance. To address these issues, we propose a low-cost real-time framework for industrial action recognition using foundation models, denoted as LRIAR, to enhance recognition accuracy and transferability while minimizing human annotation and computational overhead. The proposed framework constructs an automatically labeled dataset by coupling Grounding DINO with the pretrained BLIP-2 image encoder, enabling efficient and scalable action labeling. Leveraging the constructed dataset, we train YOLOv5 for real-time action detection, and a Vision Transformer (ViT) classifier is deceloped via LoRA-based fine-tuning for action classification. Extensive experiments conducted in real-world industrial settings validate the effectiveness of LRIAR, demonstrating consistent improvements over state-of-the-art methods in recognition accuracy, scenario generalization, and deployment efficiency.

cs.CV↗

arXiv2025

IPFormer-VideoLLM: Enhancing Multi-modal Video Understanding for Multi-shot Scenes

Video Large Language Models (VideoLLMs) have demonstrated remarkable understanding capabilities, but are found struggling to tackle multi-shot scenarios,e.g., video clips with varying camera angles or scene changes. This challenge can render failures such as instance identity forgetting and key frame negligence. In this work, we first attribute the challenge to the lack of multi-shot annotations among existing datasets and therefore we introduce a new dataset termed MultiClip-Bench, featuring dense descriptions and instruction-based question-answering pairs tailored for multi-shot scenarios. We empirically find that the training set significantly boosts the multi-shot performance, while the testing benchmark provides a reliable measure of the model capability in multi-shot scenarios. By further analyzing and discovering that current models only encode instance features in a discrete or lossy manner, at the risk of missing identity information, we then contribute a new model IPFormer-VideoLLM. Its key idea is the injection of instance-level features as instance prompts through an efficient attention-based connector. This allows for the aggregation of instance-specific information across scenes. Experiments demonstrate that our proposed dataset and model not only enhance the multi-scene video understanding significantly, but also offer distinct advantages across various video benchmarks.

cs.CV↗

arXiv2025

SRefiner: Soft-Braid Attention for Multi-Agent Trajectory Refinement

Accurate prediction of multi-agent future trajectories is crucial for autonomous driving systems to make safe and efficient decisions. Trajectory refinement has emerged as a key strategy to enhance prediction accuracy. However, existing refinement methods often overlook the topological relationships between trajectories, which are vital for improving prediction precision. Inspired by braid theory, we propose a novel trajectory refinement approach, Soft-Braid Refiner (SRefiner), guided by the soft-braid topological structure of trajectories using Soft-Braid Attention. Soft-Braid Attention captures spatio-temporal topological relationships between trajectories by considering both spatial proximity and vehicle motion states at ``soft intersection points". Additionally, we extend this approach to model interactions between trajectories and lanes, further improving the prediction accuracy. SRefiner is a multi-iteration, multi-agent framework that iteratively refines trajectories, incorporating topological information to enhance interactions within traffic scenarios. SRefiner achieves significant performance improvements over four baseline methods across two datasets, establishing a new state-of-the-art in trajectory refinement. Code is here https://github.com/Liwen-Xiao/SRefiner.

cs.RO↗

arXiv2025

FocusChat: Text-guided Long Video Understanding via Spatiotemporal Information Filtering

Recently, multi-modal large language models have made significant progress. However, visual information lacking of guidance from the user's intention may lead to redundant computation and involve unnecessary visual noise, especially in long, untrimmed videos. To address this issue, we propose FocusChat, a text-guided multi-modal large language model (LLM) that emphasizes visual information correlated to the user's prompt. In detail, Our model first undergoes the semantic extraction module, which comprises a visual semantic branch and a text semantic branch to extract image and text semantics, respectively. The two branches are combined using the Spatial-Temporal Filtering Module (STFM). STFM enables explicit spatial-level information filtering and implicit temporal-level feature filtering, ensuring that the visual tokens are closely aligned with the user's query. It lowers the essential number of visual tokens inputted into the LLM. FocusChat significantly outperforms Video-LLaMA in zero-shot experiments, using an order of magnitude less training data with only 16 visual tokens occupied. It achieves results comparable to the state-of-the-art in few-shot experiments, with only 0.72M pre-training data.

cs.CV↗

arXiv2025

PUGS: Zero-shot Physical Understanding with Gaussian Splatting

Current robotic systems can understand the categories and poses of objects well. But understanding physical properties like mass, friction, and hardness, in the wild, remains challenging. We propose a new method that reconstructs 3D objects using the Gaussian splatting representation and predicts various physical properties in a zero-shot manner. We propose two techniques during the reconstruction phase: a geometry-aware regularization loss function to improve the shape quality and a region-aware feature contrastive loss function to promote region affinity. Two other new techniques are designed during inference: a feature-based property propagation module and a volume integration module tailored for the Gaussian representation. Our framework is named as zero-shot physical understanding with Gaussian splatting, or PUGS. PUGS achieves new state-of-the-art results on the standard benchmark of ABO-500 mass prediction. We provide extensive quantitative ablations and qualitative visualization to demonstrate the mechanism of our designs. We show the proposed methodology can help address challenging real-world grasping tasks. Our codes, data, and models are available at https://github.com/EverNorif/PUGS

cs.CV↗

arXiv2025

Exploring Contextual Attribute Density in Referring Expression Counting

Referring expression counting (REC) algorithms are for more flexible and interactive counting ability across varied fine-grained text expressions. However, the requirement for fine-grained attribute understanding poses challenges for prior arts, as they struggle to accurately align attribute information with correct visual patterns. Given the proven importance of ''visual density'', it is presumed that the limitations of current REC approaches stem from an under-exploration of ''contextual attribute density'' (CAD). In the scope of REC, we define CAD as the measure of the information intensity of one certain fine-grained attribute in visual regions. To model the CAD, we propose a U-shape CAD estimator in which referring expression and multi-scale visual features from GroundingDINO can interact with each other. With additional density supervision, we can effectively encode CAD, which is subsequently decoded via a novel attention procedure with CAD-refined queries. Integrating all these contributions, our framework significantly outperforms state-of-the-art REC methods, achieves $30\%$ error reduction in counting metrics and a $10\%$ improvement in localization accuracy. The surprising results shed light on the significance of contextual attribute density for REC. Code will be at github.com/Xu3XiWang/CAD-GD.

cs.CV↗

arXiv2025

Learning and discovering multiple solutions using physics-informed neural networks with random initialization and deep ensemble

We explore the capability of physics-informed neural networks (PINNs) to discover multiple solutions. Many real-world phenomena governed by nonlinear differential equations (DEs), such as fluid flow, exhibit multiple solutions under the same conditions, yet capturing this solution multiplicity remains a significant challenge. A key difficulty is giving appropriate initial conditions or initial guesses, to which the widely used time-marching schemes and Newton's iteration method are very sensitive in finding solutions for complex computational problems. While machine learning models, particularly PINNs, have shown promise in solving DEs, their ability to capture multiple solutions remains underexplored. In this work, we propose a simple and practical approach using PINNs to learn and discover multiple solutions. We first reveal that PINNs, when combined with random initialization and deep ensemble method -- originally developed for uncertainty quantification -- can effectively uncover multiple solutions to nonlinear ordinary and partial differential equations (ODEs/PDEs). Our approach highlights the critical role of initialization in shaping solution diversity, addressing an often-overlooked aspect of machine learning for scientific computing. Furthermore, we propose utilizing PINN-generated solutions as initial conditions or initial guesses for conventional numerical solvers to enhance accuracy and efficiency in capturing multiple solutions. Extensive numerical experiments, including the Allen-Cahn equation and cavity flow, where our approach successfully identifies both stable and unstable solutions, validate the effectiveness of our method. These findings establish a general and efficient framework for addressing solution multiplicity in nonlinear differential equations.

cs.LG↗

arXiv2025

Walking with Terrain Reconstruction: Learning to Traverse Risky Sparse Footholds

Traversing risky terrains with sparse footholds presents significant challenges for legged robots, requiring precise foot placement in safe areas. To acquire comprehensive exteroceptive information, prior studies have employed motion capture systems or mapping techniques to generate heightmap for locomotion policy. However, these approaches require specialized pipelines and often introduce additional noise. While depth images from egocentric vision systems are cost-effective, their limited field of view and sparse information hinder the integration of terrain structure details into implicit features, which are essential for generating precise actions. In this paper, we demonstrate that end-to-end reinforcement learning relying solely on proprioception and depth images is capable of traversing risky terrains with high sparsity and randomness. Our method introduces local terrain reconstruction, leveraging the benefits of clear features and sufficient information from the heightmap, which serves as an intermediary for visual feature extraction and motion generation. This allows the policy to effectively represent and memorize critical terrain information. We deploy the proposed framework on a low-cost quadrupedal robot, achieving agile and adaptive locomotion across various challenging terrains and showcasing outstanding performance in real-world scenarios. Video at: youtu.be/Rj9v5EZsn-M.

cs.RO↗

arXiv2024

Subconscious Robotic Imitation Learning

Although robotic imitation learning (RIL) is promising for embodied intelligent robots, existing RIL approaches rely on computationally intensive multi-model trajectory predictions, resulting in slow execution and limited real-time responsiveness. Instead, human beings subconscious can constantly process and store vast amounts of information from their experiences, perceptions, and learning, allowing them to fulfill complex actions such as riding a bike, without consciously thinking about each. Inspired by this phenomenon in action neurology, we introduced subconscious robotic imitation learning (SRIL), wherein cognitive offloading was combined with historical action chunkings to reduce delays caused by model inferences, thereby accelerating task execution. This process was further enhanced by subconscious downsampling and pattern augmented learning policy wherein intent-rich information was addressed with quantized sampling techniques to improve manipulation efficiency. Experimental results demonstrated that execution speeds of the SRIL were 100\% to 200\% faster over SOTA policies for comprehensive dual-arm tasks, with consistently higher success rates.

cs.RO↗