arXiv subjects

Wenbin Li

Publications and source records attributed to Wenbin Li.

At least 55 records · Page 3Linked to original sources

arXiv2026

Branch, or Layer? Zeroth-Order Optimization for Continual Learning of Vision-Language Models

Vision-Language Continual Learning (VLCL) has attracted significant research attention for its robust capabilities, and the adoption of Parameter-Efficient Fine-Tuning (PEFT) strategies is enabling these models to achieve competitive performance with substantially reduced resource consumption. However, dominated First-Order (FO) optimization is prone to trap models in suboptimal local minima, especially in limited exploration subspace within PEFT. To overcome this challenge, this paper pioneers a systematic exploration of adopting Zeroth-Order (ZO) optimization for PEFT-based VLCL. We first identify the incompatibility of naive full-ZO adoption in VLCL due to optimization process instability. We then investigate the application of ZO optimization from a modality branch-wise to a fine-grained layer-wise across various training units to identify an optimal strategy. Besides, a key theoretical insight reveals that vision modality exhibit higher variance than language counterparts in VLCL during the ZO optimization process, and we propose a modality-aware ZO strategy, which adopts gradient sign normalization in ZO and constrains vision modality perturbation to further improve performance. Benefiting from the adoption of ZO optimization, PEFT-based VLCL fulfills better ability to escape local minima during the optimization process, extensive experiments on four benchmarks demonstrate that our method achieves state-of-the-art results.

cs.CV↗

arXiv2026

Training-Free Video Editing via Optical Flow-Enhanced Score Distillation

The rapid advancement in visual generation, particularly the emergence of pre-trained text-to-image and text-to-video models, has catalyzed growing interest in training-free video editing research. Mirroring training-free image editing techniques, current approaches preserve original video information through video input inversion and manipulating intermediate features and attention during the inference process to achieve content editing. Although they have demonstrated promising results, the lossy nature of the inversion process poses significant challenges in maintaining unedited regions of the video. Furthermore, feature and attention manipulation during inference can lead to unintended over-editing and face challenges in both local temporal continuity and global content consistency. To address these challenges, this study proposes a score distillation paradigm based on pre-trained text-to-video models, where the original video is iteratively optimized through multiple steps guided by editing gradients provided by score distillation to ultimately obtain the target video. The iterative optimization starting from the original video, combined with content preservation loss, ensures the maintenance of unedited regions in the original video and suppresses over-editing. To further guarantee video content consistency and temporal continuity, we additionally introduce a global consistency auxiliary loss and optical flow prediction-based local editing gradient smoothing. Experiments demonstrate that these strategies effectively address the aforementioned challenges, achieving comparable or superior performance across multiple dimensions including preservation of unedited regions, local temporal continuity, and global content consistency of editing results, compared to state-of-the-art methods.

cs.CV↗

arXiv2025

Adapting In-Domain Few-Shot Segmentation to New Domains without Source Domain Retraining

Cross-domain few-shot segmentation (CD-FSS) aims to segment objects of novel classes in new domains, which is often challenging due to the diverse characteristics of target domains and the limited availability of support data. Most CD-FSS methods redesign and retrain in-domain FSS models using abundant base data from the source domain, which are effective but costly to train. To address these issues, we propose adapting informative model structures of the well-trained FSS model for target domains by learning domain characteristics from few-shot labeled support samples during inference, thereby eliminating the need for source domain retraining. Specifically, we first adaptively identify domain-specific model structures by measuring parameter importance using a novel structure Fisher score in a data-dependent manner. Then, we progressively train the selected informative model structures with hierarchically constructed training samples, progressing from fewer to more support shots. The resulting Informative Structure Adaptation (ISA) method effectively addresses domain shifts and equips existing well-trained in-domain FSS models with flexible adaptation capabilities for new domains, eliminating the need to redesign or retrain CD-FSS models on base data. Extensive experiments validate the effectiveness of our method, demonstrating superior performance across multiple CD-FSS benchmarks. Codes are at https://github.com/fanq15/ISA.

cs.CV↗

arXiv2025

FUSE-RSVLM: Feature Fusion Vision-Language Model for Remote Sensing

Large vision-language models (VLMs) exhibit strong performance across various tasks. However, these VLMs encounter significant challenges when applied to the remote sensing domain due to the inherent differences between remote sensing images and natural images. Existing remote sensing VLMs often fail to extract fine-grained visual features and suffer from visual forgetting during deep language processing. To address this, we introduce MF-RSVLM, a Multi-Feature Fusion Remote Sensing Vision--Language Model that effectively extracts and fuses visual features for RS understanding. MF-RSVLM learns multi-scale visual representations and combines global context with local details, improving the capture of small and complex structures in RS scenes. A recurrent visual feature injection scheme ensures the language model remains grounded in visual evidence and reduces visual forgetting during generation. Extensive experiments on diverse RS benchmarks show that MF-RSVLM achieves state-of-the-art or highly competitive performance across remote sensing classification, image captioning, and VQA tasks. Our code is publicly available at https://github.com/Yunkaidang/RSVLM.

cs.CV↗

arXiv2025

LibContinual: A Comprehensive Library towards Realistic Continual Learning

A fundamental challenge in Continual Learning (CL) is catastrophic forgetting, where adapting to new tasks degrades the performance on previous ones. While the field has evolved with diverse methods, this rapid surge in diverse methodologies has culminated in a fragmented research landscape. The lack of a unified framework, including inconsistent implementations, conflicting dependencies, and varying evaluation protocols, makes fair comparison and reproducible research increasingly difficult. To address this challenge, we propose LibContinual, a comprehensive and reproducible library designed to serve as a foundational platform for realistic CL. Built upon a high-cohesion, low-coupling modular architecture, LibContinual integrates 19 representative algorithms across five major methodological categories, providing a standardized execution environment. Meanwhile, leveraging this unified framework, we systematically identify and investigate three implicit assumptions prevalent in mainstream evaluation: (1) offline data accessibility, (2) unregulated memory resources, and (3) intra-task semantic homogeneity. We argue that these assumptions often overestimate the real-world applicability of CL methods. Through our comprehensive analysis using strict online CL settings, a novel unified memory budget protocol, and a proposed category-randomized setting, we reveal significant performance drops in many representative CL methods when subjected to these real-world constraints. Our study underscores the necessity of resource-aware and semantically robust CL strategies, and offers LibContinual as a foundational toolkit for future research in realistic continual learning. The source code is available from \href{https://github.com/RL-VIG/LibContinual}{https://github.com/RL-VIG/LibContinual}.

cs.LG↗

arXiv2025

A Benchmark for Ultra-High-Resolution Remote Sensing MLLMs

Multimodal large language models (MLLMs) demonstrate strong perception and reasoning performance on existing remote sensing (RS) benchmarks. However, most prior benchmarks rely on low-resolution imagery, and some high-resolution benchmarks suffer from flawed reasoning-task designs. We show that text-only LLMs can perform competitively with multimodal vision-language models on RS reasoning tasks without access to images, revealing a critical mismatch between current benchmarks and the intended evaluation of visual understanding. To enable faithful assessment, we introduce RSHR-Bench, a super-high-resolution benchmark for RS visual understanding and reasoning. RSHR-Bench contains 5,329 full-scene images with a long side of at least 4,000 pixels, with up to about 3 x 10^8 pixels per image, sourced from widely used RS corpora and UAV collections. We design four task families: multiple-choice VQA, open-ended VQA, image captioning, and single-image evaluation. These tasks cover nine perception categories and four reasoning types, supporting multi-turn and multi-image dialog. To reduce reliance on language priors, we apply adversarial filtering with strong LLMs followed by rigorous human verification. Overall, we construct 3,864 VQA tasks, 3,913 image captioning tasks, and 500 fully human-written or verified single-image evaluation VQA pairs. Evaluations across open-source, closed-source, and RS-specific VLMs reveal persistent performance gaps in super-high-resolution scenarios. Code: https://github.com/Yunkaidang/RSHR

cs.CV↗

arXiv2025

ResDynUNet++: A nested U-Net with residual dynamic convolution blocks for dual-spectral CT

We propose a hybrid reconstruction framework for dual-spectral CT (DSCT) that integrates iterative methods with deep learning models. The reconstruction process consists of two complementary components: a knowledge-driven module and a data-driven module. In the knowledge-driven phase, we employ the oblique projection modification technique (OPMT) to reconstruct an intermediate solution of the basis material images from the projection data. We select OPMT for this role because of its fast convergence, which allows it to rapidly generate an intermediate solution that successfully achieves basis material decomposition. Subsequently, in the data-driven phase, we introduce a novel neural network, ResDynUNet++, to refine this intermediate solution. The ResDynUNet++ is built upon a UNet++ backbone by replacing standard convolutions with residual dynamic convolution blocks, which combine the adaptive, input-specific feature extraction of dynamic convolution with the stable training of residual connections. This architecture is designed to address challenges like channel imbalance and near-interface large artifacts in DSCT, producing clean and accurate final solutions. Extensive experiments on both synthetic phantoms and real clinical datasets validate the efficacy and superior performance of the proposed method.

cs.CV↗

arXiv2025

SEA: Semantic Map Prediction for Active Exploration of Uncertain Areas

In this paper, we propose SEA, a novel approach for active robot exploration through semantic map prediction and a reinforcement learning-based hierarchical exploration policy. Unlike existing learning-based methods that rely on one-step waypoint prediction, our approach enhances the agent's long-term environmental understanding to facilitate more efficient exploration. We propose an iterative prediction-exploration framework that explicitly predicts the missing areas of the map based on current observations. The difference between the actual accumulated map and the predicted global map is then used to guide exploration. Additionally, we design a novel reward mechanism that leverages reinforcement learning to update the long-term exploration strategies, enabling us to construct an accurate semantic map within limited steps. Experimental results demonstrate that our method significantly outperforms state-of-the-art exploration strategies, achieving superior coverage ares of the global map within the same time constraints.

cs.RO↗

arXiv2025

A level-set structural approach for multi-physics joint inversion using full-waveform and gravity data

This paper presents a level-set based structural approach for the joint inversion of full-waveform and gravity data. The joint inversion aims to integrate the strengths of full-waveform inversion for high resolution imaging and gravity inversion for detecting density contrasts over extensive regions. Although common studies typically only observe full-waveform inversion assisting gravity inversion, we propose three key points that enable gravity data to complement full-waveform data in the joint inversion. (i) Based on the well-posedness theorem, we consider a volume mass distribution where the density-contrast value is imposed as a priori information, ensuring that the gravity data provide meaningful information. (ii) We utilize a level-set formulation to characterize the shared interface of wave velocity and density functions, connecting multi-physics datasets via the structural similarity of their inversion parameters. (iii) We develop a balanced and decaying weight to regulate the influence of multi-physics datasets during joint inversion. This weight comprises a balanced part that accounts for the differing scales of full-waveform and gravity data, and a decaying part designed to effectively utilize the features and advantages of each dataset.

physics.geo-ph↗

arXiv2025

Microscopic Theory of Light-Induced Coherent Phonons Mediated by Quantum Geometry

Light-induced coherent phonons provide a powerful platform for ultrafast control of material properties. However, the microscopic theory and quantum geometric nature of this phenomenon remain underexplored. Here, we develop a fully quantum-mechanical framework based on Feynman diagrams to systematically describe the generation of coherent phonons by light. We identify a dominant second-order, double-resonant process in noncentrosymmetric semiconductors that efficiently couples light to both electronic and phononic excitations. Crucially, we uncover the quantum geometric origin, encoded in the electron-phonon coupling (EPC) shift vector and the EPC quantum geometric tensor. Applying our theory to ferroelectric BaTiO$_3$ and SnSe, we demonstrate the potential for light-induced modulation of ferroelectric polarization driven by coherent phonons. This work provides fundamental insights for designing efficient optical control strategies for both coherent phonons and ferroelectric polarization.

cond-mat.mes-hall↗

arXiv2025

Learning Dolly-In Filming From Demonstration Using a Ground-Based Robot

Cinematic camera control demands a balance of precision and artistry - qualities that are difficult to encode through handcrafted reward functions. While reinforcement learning (RL) has been applied to robotic filmmaking, its reliance on bespoke rewards and extensive tuning limits creative usability. We propose a Learning from Demonstration (LfD) approach using Generative Adversarial Imitation Learning (GAIL) to automate dolly-in shots with a free-roaming, ground-based filming robot. Expert trajectories are collected via joystick teleoperation in simulation, capturing smooth, expressive motion without explicit objective design. Trained exclusively on these demonstrations, our GAIL policy outperforms a PPO baseline in simulation, achieving higher rewards, faster convergence, and lower variance. Crucially, it transfers directly to a real-world robot without fine-tuning, achieving more consistent framing and subject alignment than a prior TD3-based method. These results show that LfD offers a robust, reward-free alternative to RL in cinematic domains, enabling real-time deployment with minimal technical effort. Our pipeline brings intuitive, stylized camera control within reach of creative professionals, bridging the gap between artistic intent and robotic autonomy.

cs.RO↗

arXiv2025

Reinforcement Learning of Dolly-In Filming Using a Ground-Based Robot

Free-roaming dollies enhance filmmaking with dynamic movement, but challenges in automated camera control remain unresolved. Our study advances this field by applying Reinforcement Learning (RL) to automate dolly-in shots using free-roaming ground-based filming robots, overcoming traditional control hurdles. We demonstrate the effectiveness of combined control for precise film tasks by comparing it to independent control strategies. Our robust RL pipeline surpasses traditional Proportional-Derivative controller performance in simulation and proves its efficacy in real-world tests on a modified ROSBot 2.0 platform equipped with a camera turret. This validates our approach's practicality and sets the stage for further research in complex filming scenarios, contributing significantly to the fusion of technology with cinematic creativity. This work presents a leap forward in the field and opens new avenues for research and development, effectively bridging the gap between technological advancement and creative filmmaking.

cs.RO↗

arXiv2025

GTool: Graph Enhanced Tool Planning with Large Language Model

Tool planning with large language models (LLMs), referring to selecting, organizing, and preparing the tools necessary to complete a user request, bridges the gap between natural language understanding and task execution. However, current works treat different tools as isolated components and fail to leverage the inherent dependencies of tools, leading to invalid planning results. Since tool dependencies are often incomplete, it becomes challenging for LLMs to accurately identify the appropriate tools required by a user request, especially when confronted with a large toolset. To solve this challenge, we propose \texttt{GTool}, which is the first work aiming to enhance the tool planning ability of LLMs under incomplete dependencies. \texttt{GTool} constructs a request-specific tool graph to select tools efficiently and generate the \texttt{ } which provides sufficient dependency information understandable by LLMs. Moreover, a missing dependency prediction task is designed to improve the reliability of \texttt{GTool} with incomplete dependencies. Without trimming LLMs, \texttt{GTool} can be seamlessly integrated with various LLM backbones without extensive retraining. Extensive experiments show that \texttt{GTool} achieves more than 29.6\% performance improvements compared with the state-of-the-art (SOTA) baselines with a light-weight (7B) LLM backbone.

cs.AI↗

arXiv2025

Experimental Realization of the Topologically Nontrivial Phase in Monolayer Si$_2$Te$_2$

The free-standing monolayer Si$_2$Te$_2$ (ML-Si$_2$Te$_2$) has been theoretically predicted to host a room-temperature quantum spin Hall phase. However, its experimental realization remains challenge due to the absence of a three-dimensional counterpart. Here, we demonstrate that HfTe$_2$ serves as an ideal substrate for the epitaxial growth of ML-Si$_2$Te$_2$, preserving its topological phase. Scanning tunneling microscopy and spectroscopy confirm a strain-free ${(1 \times 1)}$ lattice of ML-Si$_2$Te$_2$, along with a sizable band gap, which is well captured by first-principles calculations. Moreover, distinct edge states, independent of step geometry and exhibiting a broad spatial distribution, are observed at ML-Si$_2$Te$_2$ step edges, underscoring its topological nature.

cond-mat.mtrl-sci↗

arXiv2025

Vortex Pinning in Niobium covered by a thin polycrystalline Gold

Owing to its superconducting properties, Niobium (Nb) is an excellent candidate material for superconducting electronics and applications in quantum technology. Here we perform scanning tunneling microscopy and spectroscopy experiments on Nb films covered by a thin gold (Au) film. We investigate the minigap structure of the proximitized region and provide evidence for a highly transparent interface between Nb and Au, beneficial for device applications. Imaging of Abrikosov vortices in presence of a perpendicular magnetic field is reported. The data show vortex pinning by the granular structure of the polycrystalline Au film. Our results show robust and homogeneous superconducting properties of thin Nb film in the presence of a gold capping layer. The Au film not only protects the Nb from surface oxidation but also preserves its excellent superconducting properties.

cond-mat.supr-con↗

arXiv2025

RoundaboutHD: High-Resolution Real-World Urban Environment Benchmark for Multi-Camera Vehicle Tracking

The multi-camera vehicle tracking (MCVT) framework holds significant potential for smart city applications, including anomaly detection, traffic density estimation, and suspect vehicle tracking. However, current publicly available datasets exhibit limitations, such as overly simplistic scenarios, low-resolution footage, and insufficiently diverse conditions, creating a considerable gap between academic research and real-world scenario. To fill this gap, we introduce RoundaboutHD, a comprehensive, high-resolution multi-camera vehicle tracking benchmark dataset specifically designed to represent real-world roundabout scenarios. RoundaboutHD provides a total of 40 minutes of labelled video footage captured by four non-overlapping, high-resolution (4K resolution, 15 fps) cameras. In total, 512 unique vehicle identities are annotated across different camera views, offering rich cross-camera association data. RoundaboutHD offers temporal consistency video footage and enhanced challenges, including increased occlusions and nonlinear movement inside the roundabout. In addition to the full MCVT dataset, several subsets are also available for object detection, single camera tracking, and image-based vehicle re-identification (ReID) tasks. Vehicle model information and camera modelling/ geometry information are also included to support further analysis. We provide baseline results for vehicle detection, single-camera tracking, image-based vehicle re-identification, and multi-camera tracking. The dataset and the evaluation code are publicly available at: https://github.com/siri-rouser/RoundaboutHD.git

cs.CV↗

arXiv2025

Ferroelectricity in 6 Angstrom-Thick Two-dimensional Ga$_2$O$_3$

Atomic-scale ferroelectric thin films hold great promise for high-density, low-power applications but face stability and voltage scaling challenges at extreme thinness. Here, we demonstrate ferroelectricity in single-crystalline two-dimensional (2D) Ga$_2$O$_3$, an ultra-wide-bandgap semiconductor, at just 6 angstrom thickness, exhibiting exceptional retention and thermal stability. We show that epitaxial beta-Ga$_2$O$_3$ can be exfoliated down to a half-unit cell thickness via a self-limiting mechanism, enabling a biaxial strain-induced phase transition into a novel ferroelectric layered structure. Strain modulation enables the reduction of polarization switching voltage to 0.8 V, meeting CMOS voltage scaling requirements. Theoretical calculations reveal that switching is driven by covalent bond reconstruction, effectively countering depolarization and enhancing stability. Additionally, we integrate ferroelectric 2D Ga$_2$O$_3$ onto silicon using a low-temperature, back-end-of-line-compatible process. This work advances the exploration of sub-nanometer ferroelectrics, paving the way for high-density, low-power, non-volatile applications seamlessly integrated with advanced silicon technology.

cond-mat.mtrl-sci↗

arXiv2025

Optical Manifestations of Quantum Geometry in Electron-Phonon Coupling

Quantum geometry is crucial for understanding intricate condensed matter systems, governing transport phenomena and optical responses. However, traditional studies predominantly consider a static crystal lattice, focusing exclusively on the pure-electronic quantum geometry of the Hilbert space parameterized by electronic wave vectors, thereby overlooking the dynamic effects arising from phonons and their coupling with electrons. In this work, we reveal the intrinsic quantum geometry of the electron-phonon coupling (EPC), which resides in the hybrid Hilbert space parameterized by both the electronic wave vectors and phonon displacements. The EPC quantum metric, EPC Berry curvature and EPC shift vector, as central elements, quantify the EPC-induced velocity, polarization, and anomalous charge-center shift, respectively. We further connect this geometry to phonon-mediated optical responses, particularly in-gap resonances, enabling experimental detection and characterization of EPC quantum geometry.

cond-mat.mes-hall↗