Searcharxiv⌕ Search

arXiv subjects

Yiming Wang

Publications and source records attributed to Yiming Wang.

At least 91 records · Page 5Linked to original sources
arXiv2026

StructXLIP: Enhancing Vision-language Models with Multimodal Structural Cues

Edge-based representations are fundamental cues for visual understanding, a principle rooted in early vision research and still central today. We extend this principle to vision-language alignment, showing that isolating and aligning structural cues across modalities can greatly benefit fine-tuning on long, detail-rich captions, with a specific focus on improving cross-modal retrieval. We introduce StructXLIP, a fine-tuning alignment paradigm that extracts edge maps (e.g., Canny), treating them as proxies for the visual structure of an image, and filters the corresponding captions to emphasize structural cues, making them "structure-centric". Fine-tuning augments the standard alignment loss with three structure-centric losses: (i) aligning edge maps with structural text, (ii) matching local edge regions to textual chunks, and (iii) connecting edge maps to color images to prevent representation drift. From a theoretical standpoint, while standard CLIP maximizes the mutual information between visual and textual embeddings, StructXLIP additionally maximizes the mutual information between multimodal structural representations. This auxiliary optimization is intrinsically harder, guiding the model toward more robust and semantically stable minima, enhancing vision-language alignment. Beyond outperforming current competitors on cross-modal retrieval in both general and specialized domains, our method serves as a general boosting recipe that can be integrated into future approaches in a plug-and-play manner. Code and pretrained models are publicly available at: https://github.com/intelligolabs/StructXLIP.

cs.CV↗
arXiv2026

Multi-Level Conditioning by Pairing Localized Text and Sketch for Fashion Image Generation

Sketches offer designers a concise yet expressive medium for early-stage fashion ideation by specifying structure, silhouette, and spatial relationships, while textual descriptions complement sketches to convey material, color, and stylistic details. Effectively combining textual and visual modalities requires adherence to the sketch visual structure when leveraging the guidance of localized attributes from text. We present LOcalized Text and Sketch with multi-level guidance (LOTS), a framework that enhances fashion image generation by combining global sketch guidance with multiple localized sketch-text pairs. LOTS employs a Multi-level Conditioning Stage to independently encode local features within a shared latent space while maintaining global structural coordination. Then, the Diffusion Pair Guidance stage integrates both local and global conditioning via attention-based guidance within the diffusion model's multi-step denoising process. To validate our method, we develop Sketchy, the first fashion dataset where multiple text-sketch pairs are provided per image. Sketchy provides high-quality, clean sketches with a professional look and consistent structure. To assess robustness beyond this setting, we also include an "in the wild" split with non-expert sketches, featuring higher variability and imperfections. Experiments demonstrate that our method strengthens global structural adherence while leveraging richer localized semantic guidance, achieving improvement over state-of-the-art. The dataset, platform, and code are publicly available.

cs.CV↗
arXiv2026

ExoGS: A 4D Real-to-Sim-to-Real Framework for Scalable Manipulation Data Collection

Real-to-Sim-to-Real technique is gaining increasing interest for robotic manipulation, as it can generate scalable data in simulation while having narrower sim-to-real gap. However, previous methods mainly focused on environment-level visual real-to-sim transfer, ignoring the transfer of interactions, which could be challenging and inefficient to obtain purely in simulation especially for contact-rich tasks. We propose ExoGS, a robot-free 4D Real-to-Sim-to-Real framework that captures both static environments and dynamic interactions in the real world and transfers them seamlessly to a simulated environment. It provides a new solution for scalable manipulation data collection and policy learning. ExoGS employs a self-designed robot-isomorphic passive exoskeleton AirExo-3 to capture kinematically consistent trajectories with millimeter-level accuracy and synchronized RGB observations during direct human demonstrations. The robot, objects, and environment are reconstructed as editable 3D Gaussian Splatting assets, enabling geometry-consistent replay and large-scale data augmentation. Additionally, a lightweight Mask Adapter injects instance-level semantics into the policy to enhance robustness under visual domain shifts. Real-world experiments demonstrate that ExoGS significantly improves data efficiency and policy generalization compared to teleoperation-based baselines. Code and hardware files have been released on https://github.com/zaixiabalala/ExoGS.

cs.RO↗
arXiv2026

Semantic-Guided Unsupervised Video Summarization

Video summarization is a crucial technique for social understanding, enabling efficient browsing of massive multimedia content and extraction of key information from social platforms. Most existing unsupervised summarization methods rely on Generative Adversarial Networks (GANs) to enhance keyframe selection and generate coherent, video summaries through adversarial training. However, such approaches primarily exploit unimodal features, overlooking the guiding role of semantic information in keyframe selection, and often suffer from unstable training. To address these limitations, we propose a novel Semantic-Guided Unsupervised Video Summarization method. Specifically, we design a novel frame-level semantic alignment attention mechanism and integrate it into a keyframe selector, which guides the Transformer-based generator within the adversarial framework to better reconstruct videos. In addition, we adopt an incremental training strategy to progressively update the model components, effectively mitigating the instability of GAN training. Experimental results demonstrate that our approach achieves superior performance on multiple benchmark datasets.

cs.AI↗
arXiv2026

Intelligent Distributed Optical Fiber Sensing in Transportation Infrastructures: Research Progress, Applications, and Challenges

Distributed optical fiber sensing (DOFS), along with its capabilities of long-range coverage, multi-parameter monitoring, and completely passive detection, emerges as one of the most promising non-destructive detection techniques for structural health monitoring (SHM) and operational assessment of linear transportation infrastructures. In this paper, we provide a state-of-the-art review on DOFS applications across typical linear infrastructure systems, encompassing highways, long-span bridges, rail transit networks, airport runways, and analogous linear structures. The comprehensive discussion consists of four critical research dimensions: 1) optical fiber selection for multi-parameter sensing and robust cable packaging techniques, 2) distributed sensing principles and signal processing algorithms, 3) diverse application scenarios in SHM and related fields, and 4) anomaly detection and event classification methodologies. Building upon the foundational introduction of DOFS technical principles and monitoring solutions for intelligent transportation infrastructure, this paper elaborates on system design approaches, sensing data analytics algorithms, and future research directions.

physics.optics↗
arXiv2026

Enhancing Retrieval Augmentation via Adversarial Collaboration

Retrieval-augmented Generation (RAG) is a prevalent approach for domain-specific LLMs, yet it is often plagued by "Retrieval Hallucinations"--a phenomenon where fine-tuned models fail to recognize and act upon poor-quality retrieved documents, thus undermining performance. To address this, we propose the Adversarial Collaboration RAG (AC-RAG) framework. AC-RAG employs two heterogeneous agents: a generalist Detector that identifies knowledge gaps, and a domain-specialized Resolver that provides precise solutions. Guided by a moderator, these agents engage in an adversarial collaboration, where the Detector's persistent questioning challenges the Resolver's expertise. This dynamic process allows for iterative problem dissection and refined knowledge retrieval. Extensive experiments show that AC-RAG significantly improves retrieval accuracy and outperforms state-of-the-art RAG methods across various vertical domains.

cs.AI↗
arXiv2026

Spherical Geometry Diffusion: Generating High-quality 3D Face Geometry via Sphere-anchored Representations

A fundamental challenge in text-to-3D face generation is achieving high-quality geometry. The core difficulty lies in the arbitrary and intricate distribution of vertices in 3D space, making it challenging for existing models to establish clean connectivity and resulting in suboptimal geometry. To address this, our core insight is to simplify the underlying geometric structure by constraining the distribution onto a simple and regular manifold, a topological sphere. Building on this, we first propose the Spherical Geometry Representation, a novel face representation that anchors geometric signals to uniform spherical coordinates. This guarantees a regular point distribution, from which the mesh connectivity can be robustly reconstructed. Critically, this canonical sphere can be seamlessly unwrapped into a 2D map, creating a perfect synergy with powerful 2D generative models. We then introduce Spherical Geometry Diffusion, a conditional diffusion framework built upon this 2D map. It enables diverse and controllable generation by jointly modeling geometry and texture, where the geometry explicitly conditions the texture synthesis process. Our method's effectiveness is demonstrated through its success in a wide range of tasks: text-to-3D generation, face reconstruction, and text-based 3D editing. Extensive experiments show that our approach substantially outperforms existing methods in geometric quality, textual fidelity, and inference efficiency.

cs.CV↗
arXiv2026

Unifying Speech Recognition, Synthesis and Conversion with Autoregressive Transformers

Traditional speech systems typically rely on separate, task-specific models for text-to-speech (TTS), automatic speech recognition (ASR), and voice conversion (VC), resulting in fragmented pipelines that limit scalability, efficiency, and cross-task generalization. In this paper, we present General-Purpose Audio (GPA), a unified audio foundation model that integrates multiple core speech tasks within a single large language model (LLM) architecture. GPA operates on a shared discrete audio token space and supports instruction-driven task induction, enabling a single autoregressive model to flexibly perform TTS, ASR, and VC without architectural modifications. This unified design combines a fully autoregressive formulation over discrete speech tokens, joint multi-task training across speech domains, and a scalable inference pipeline that achieves high concurrency and throughput. The resulting model family supports efficient multi-scale deployment, including a lightweight 0.3B-parameter variant optimized for edge and resource-constrained environments. Together, these design choices demonstrate that a unified autoregressive architecture can achieve competitive performance across diverse speech tasks while remaining viable for low-latency, practical deployment.

cs.SD↗
arXiv2026

PALM-Bench: A Comprehensive Benchmark for Personalized Audio-Language Models

Large Audio-Language Models (LALMs) have demonstrated strong performance in audio understanding and generation. Yet, our extensive benchmarking reveals that their behavior is largely generic (e.g., summarizing spoken content) and fails to adequately support personalized question answering (e.g., summarizing what my best friend says). In contrast, human conditions their interpretation and decision-making on each individual's personal context. To bridge this gap, we formalize the task of Personalized LALMs (PALM) for recognizing personal concepts and reasoning within personal context. Moreover, we create the first benchmark (PALM-Bench) to foster the methodological advances in PALM and enable structured evaluation on several tasks across multi-speaker scenarios. Our extensive experiments on representative open-source LALMs, show that existing training-free prompting and supervised fine-tuning strategies, while yield improvements, remains limited in modeling personalized knowledge and transferring them across tasks robustly. Data and code will be released.

cs.CL↗
arXiv2025

Perturbative Kondo destruction and global phase diagram of heavy fermion metals

Strange metals represent a foundational problem in quantum condensed matter physics, and heavy fermion systems provide a canonical setting to advance a general understanding. The concept of a Kondo destruction quantum critical point is widely invoked to describe the competition of the Kondo effect and the local-moment magnetism. Here, we develop a unified field-theoretic approach, analyzing this competition from a rare approach that is anchored by the magnetically ordered side. Our analysis reveals, for the first time within a renormalization group framework, a quantum critical point across which the Kondo effect goes from being destroyed to dominating. Our findings elucidate not only the Kondo destruction quantum criticality but also an accompanying global phase diagram of heavy fermion metals.

cond-mat.str-el↗
arXiv2025

RGMP: Recurrent Geometric-prior Multimodal Policy for Generalizable Humanoid Robot Manipulation

Humanoid robots exhibit significant potential in executing diverse human-level skills. However, current research predominantly relies on data-driven approaches that necessitate extensive training datasets to achieve robust multimodal decision-making capabilities and generalizable visuomotor control. These methods raise concerns due to the neglect of geometric reasoning in unseen scenarios and the inefficient modeling of robot-target relationships within the training data, resulting in significant waste of training resources. To address these limitations, we present the Recurrent Geometric-prior Multimodal Policy (RGMP), an end-to-end framework that unifies geometric-semantic skill reasoning with data-efficient visuomotor control. For perception capabilities, we propose the Geometric-prior Skill Selector, which infuses geometric inductive biases into a vision language model, producing adaptive skill sequences for unseen scenes with minimal spatial common sense tuning. To achieve data-efficient robotic motion synthesis, we introduce the Adaptive Recursive Gaussian Network, which parameterizes robot-object interactions as a compact hierarchy of Gaussian processes that recursively encode multi-scale spatial relationships, yielding dexterous, data-efficient motion synthesis even from sparse demonstrations. Evaluated on both our humanoid robot and desktop dual-arm robot, the RGMP framework achieves 87% task success in generalization tests and exhibits 5x greater data efficiency than the state-of-the-art model. This performance underscores its superior cross-domain generalization, enabled by geometric-semantic reasoning and recursive-Gaussion adaptation.

cs.RO↗
arXiv2025

3DSGrasp: 3D Shape-Completion for Robotic Grasp

Real-world robotic grasping can be done robustly if a complete 3D Point Cloud Data (PCD) of an object is available. However, in practice, PCDs are often incomplete when objects are viewed from few and sparse viewpoints before the grasping action, leading to the generation of wrong or inaccurate grasp poses. We propose a novel grasping strategy, named 3DSGrasp, that predicts the missing geometry from the partial PCD to produce reliable grasp poses. Our proposed PCD completion network is a Transformer-based encoder-decoder network with an Offset-Attention layer. Our network is inherently invariant to the object pose and point's permutation, which generates PCDs that are geometrically consistent and completed properly. Experiments on a wide range of partial PCD show that 3DSGrasp outperforms the best state-of-the-art method on PCD completion tasks and largely improves the grasping success rate in real-world scenarios. The code and dataset will be made available upon acceptance.

cs.RO↗
arXiv2025

Obstruction reasoning for robotic grasping

Successful robotic grasping in cluttered environments not only requires a model to visually ground a target object but also to reason about obstructions that must be cleared beforehand. While current vision-language embodied reasoning models show emergent spatial understanding, they remain limited in terms of obstruction reasoning and accessibility planning. To bridge this gap, we present UNOGrasp, a learning-based vision-language model capable of performing visually-grounded obstruction reasoning to infer the sequence of actions needed to unobstruct the path and grasp the target object. We devise a novel multi-step reasoning process based on obstruction paths originated by the target object. We anchor each reasoning step with obstruction-aware visual cues to incentivize reasoning capability. UNOGrasp combines supervised and reinforcement finetuning through verifiable reasoning rewards. Moreover, we construct UNOBench, a large-scale dataset for both training and benchmarking, based on MetaGraspNetV2, with over 100k obstruction paths annotated by humans with obstruction ratios, contact points, and natural-language instructions. Extensive experiments and real-robot evaluations show that UNOGrasp significantly improves obstruction reasoning and grasp success across both synthetic and real-world environments, outperforming generalist and proprietary alternatives. Project website: https://tev-fbk.github.io/UnoGrasp/.

cs.RO↗
arXiv2025

Learning Efficient Fuse-and-Refine for Feed-Forward 3D Gaussian Splatting

Recent advances in feed-forward 3D Gaussian Splatting have led to rapid improvements in efficient scene reconstruction from sparse views. However, most existing approaches construct Gaussian primitives directly aligned with the pixels in one or more of the input images. This leads to redundancies in the representation when input views overlap and constrains the position of the primitives to lie along the input rays without full flexibility in 3D space. Moreover, these pixel-aligned approaches do not naturally generalize to dynamic scenes, where effectively leveraging temporal information requires resolving both redundant and newly appearing content across frames. To address these limitations, we introduce a novel Fuse-and-Refine module that enhances existing feed-forward models by merging and refining the primitives in a canonical 3D space. At the core of our method is an efficient hybrid Splat-Voxel representation: from an initial set of pixel-aligned Gaussian primitives, we aggregate local features into a coarse-to-fine voxel hierarchy, and then use a sparse voxel transformer to process these voxel features and generate refined Gaussian primitives. By fusing and refining an arbitrary number of inputs into a consistent set of primitives, our representation effectively reduces redundancy and naturally adapts to temporal frames, enabling history-aware online reconstruction of dynamic scenes. Our approach achieves state-of-the-art performance in both static and streaming scene reconstructions while running at interactive rates (15 fps with 350ms delay) on a single H100 GPU.

cs.CV↗
arXiv2025

Neural Texture Splatting: Expressive 3D Gaussian Splatting for View Synthesis, Geometry, and Dynamic Reconstruction

3D Gaussian Splatting (3DGS) has emerged as a leading approach for high-quality novel view synthesis, with numerous variants extending its applicability to a broad spectrum of 3D and 4D scene reconstruction tasks. Despite its success, the representational capacity of 3DGS remains limited by the use of 3D Gaussian kernels to model local variations. Recent works have proposed to augment 3DGS with additional per-primitive capacity, such as per-splat textures, to enhance its expressiveness. However, these per-splat texture approaches primarily target dense novel view synthesis with a reduced number of Gaussian primitives, and their effectiveness tends to diminish when applied to more general reconstruction scenarios. In this paper, we aim to achieve concrete performance improvement over state-of-the-art 3DGS variants across a wide range of reconstruction tasks, including novel view synthesis, geometry and dynamic reconstruction, under both sparse and dense input settings. To this end, we introduce Neural Texture Splatting (NTS). At the core of our approach is a global neural field (represented as a hybrid of a tri-plane and a neural decoder) that predicts local appearance and geometric fields for each primitive. By leveraging this shared global representation that models local texture fields across primitives, we significantly reduce model size and facilitate efficient global information exchange, demonstrating strong generalization across tasks. Furthermore, our neural modeling of local texture fields introduces expressive view- and time-dependent effects, a critical aspect that existing methods fail to account for. Extensive experiments show that Neural Texture Splatting consistently improves models and achieves state-of-the-art results across multiple benchmarks.

cs.CV↗
arXiv2025

Spin Excitations and Flat Electronic Bands in a Cr-based Kagome Superconductor

In the quest for topology- and correlation-driven quantum states, kagome lattice materials have garnered significant interest for their band structures, featuring flat bands (FBs) from the quantum destructive interference of the electronic wavefunction. Tuning an FB to the chemical potential could induce electronic instabilities and emergent orders. Despite extensive studies, direct evidence of FBs tuned to the chemical potential and their role in emergent orders in bulk materials remains lacking. Using angle-resolved photoemission spectroscopy, resonant inelastic X-ray scattering, and density functional theory, we show that the low-energy structure of the Cr-based kagome metal superconductor {\Cr} is dominated by FBs at the Fermi level. We also observe low-energy magnetic excitations evolving across the low-temperature transition, largely consistent with the FB shift. Our results suggest that the low-temperature order contains a magnetic origin and that the kagome FBs may play a role in the emergence of this order.

cond-mat.str-el↗
arXiv2025

Association and Consolidation: Evolutionary Memory-Enhanced Incremental Multi-View Clustering

Incremental multi-view clustering aims to achieve stable clustering results while addressing the stability-plasticity dilemma (SPD) in view-incremental scenarios. The core challenge is that the model must have enough plasticity to quickly adapt to new data, while maintaining sufficient stability to consolidate long-term knowledge. To address this challenge, we propose a novel Evolutionary Memory-Enhanced Incremental Multi-View Clustering (EMIMC), inspired by the memory regulation mechanisms of the human brain. Specifically, we design a rapid association module to establish connections between new and historical views, thereby ensuring the plasticity required for learning new knowledge. Second, a cognitive forgetting module with a decay mechanism is introduced. By dynamically adjusting the contribution of the historical view to optimize knowledge integration. Finally, we propose a knowledge consolidation module to progressively refine short-term knowledge into stable long-term memory using temporal tensors, thereby ensuring model stability. By integrating these modules, EMIMC achieves strong knowledge retention capabilities in scenarios with growing views. Extensive experiments demonstrate that EMIMC exhibits remarkable advantages over existing state-of-the-art methods.

cs.CV↗
arXiv2025

Sampling-Efficient Test-Time Scaling: Self-Estimating the Best-of-N Sampling in Early Decoding

Test-time scaling enhances large language model performance by allocating additional compute resources during inference. Best-of-N (BoN) sampling serves as a common sampling-based scaling technique, broadening the search space in parallel to find better solutions from the model distribution. However, its cost-performance trade-off is still underexplored. Two main challenges limit the efficiency of BoN sampling: (1) Generating N full samples consumes substantial GPU memory, reducing inference capacity under limited resources. (2) Reward models add extra memory and latency overhead, and training strong reward models introduces potential training data costs. Although some studies have explored efficiency improvements, none have addressed both challenges at once. To address this gap, we propose Self-Truncation Best-of-N (ST-BoN), a decoding method that avoids fully generating all N samples and eliminates the need for reward models. It leverages early sampling consistency in the model's internal states to identify the most promising path and truncate suboptimal ones. In terms of cost, ST-BoN reduces dynamic GPU memory usage by over 80% and inference latency by 50%. In terms of cost-performance trade-off, ST-BoN achieves the same performance as Full-BoN while saving computational cost by 70%-80%, and under the same cost, it can improve accuracy by 3-4 points.

cs.CL↗