arXiv subjects

Shuxian Li

Publications and source records attributed to Shuxian Li.

6 recordsLinked to original sources

arXiv2026

AdaZoom-GUI: Adaptive Zoom-based GUI Grounding with Instruction Refinement

GUI grounding is a critical capability for vision-language models (VLMs) that enables automated interaction with graphical user interfaces by locating target elements from natural language instructions. However, grounding on GUI screenshots remains challenging due to high-resolution images, small UI elements, and ambiguous user instructions. In this work, we propose AdaZoom-GUI, an adaptive zoom-based GUI grounding framework that improves both localization accuracy and instruction understanding. Our approach introduces an instruction refinement module that rewrites natural language commands into explicit and detailed descriptions, allowing the grounding model to focus on precise element localization. In addition, we design a conditional zoom-in strategy that selectively performs a second-stage inference on predicted small elements, improving localization accuracy while avoiding unnecessary computation and context loss on simpler cases. To support this framework, we construct a high-quality GUI grounding dataset and train the grounding model using Group Relative Policy Optimization (GRPO), enabling the model to predict both click coordinates and element bounding boxes. Experiments on public benchmarks demonstrate that our method achieves state-of-the-art performance among models with comparable or even larger parameter sizes, highlighting its effectiveness for high-resolution GUI understanding and practical GUI agent deployment.

cs.CV↗

arXiv2026

Nuanced Emotion Recognition Based on a Segment-based MLLM Framework Leveraging Qwen3-Omni for AH Detection

Emotion recognition in videos is a pivotal task in affective computing, where identifying subtle psychological states such as Ambivalence and Hesitancy holds significant value for behavioral intervention and digital health. Ambivalence and Hesitancy states often manifest through cross-modal inconsistencies such as discrepancies between facial expressions, vocal tones, and textual semantics, posing a substantial challenge for automated recognition. This paper proposes a recognition framework that integrates temporal segment modeling with Multimodal Large Language Models. To address computational efficiency and token constraints in long video processing, we employ a segment-based strategy, partitioning videos into short clips with a maximum duration of 5 seconds. We leverage the Qwen3-Omni-30B-A3B model, fine-tuned on the BAH dataset using LoRA and full-parameter strategies via the MS-Swift framework, enabling the model to synergistically analyze visual and auditory signals. Experimental results demonstrate that the proposed method achieves an accuracy of 85.1% on the test set, significantly outperforming existing benchmarks and validating the superior capability of Multimodal Large Language Models in capturing complex and nuanced emotional conflicts. The code is released at https://github.com/dlnn123/A-H-Detection-with-Qwen-Omni.git.

cs.CV↗

arXiv2026

Statistical Blendshape Calculation and Analysis for Graphics Applications

With the development of virtualization and AI, real-time facial avatar animation is widely used in entertainment, office, business and other fields. Against this background, blendshapes have become a common industry animation solution because of their relative simplicity and ease of interpretation. Aiming for real-time performance and low computing resource dependence, we independently developed an accurate blendshape prediction system for low-power VR applications using a standard webcam. First, blendshape feature vectors are extracted through affine transformation and segmentation. Through further transformation and regression analysis, we were able to identify models for most blendshapes with significant predictive power. Post-processing was used to further improve response stability, including smoothing filtering and nonlinear transformations to minimize error. Experiments showed the system achieved accuracy similar to ARKit 6. Our model has low sensor/hardware requirements and realtime response with a consistent, accurate and smooth visual experience.

cs.GR↗

arXiv2025

Evolving in Tasks: Empowering the Multi-modality Large Language Model as the Computer Use Agent

Computer use agents represent an emerging area in artificial intelligence, aiming to operate computers autonomously to fulfill user tasks, attracting significant attention from both industry and academia. However, the performance of existing agents remains insufficient for practical deployment. In this paper, we propose the Self-Evolution Agent (SEA) for computer operation, alongside three core innovations in data generation, reinforcement learning, and model enhancement to develop this agent. Specifically, we first design an automatic pipeline to generate verifiable task trajectories for training. Second, we propose Efficient Step-wise Reinforcement Learning to reduce the substantial computational overhead of long-horizon training. Finally, we introduce a model enhancement method that integrates grounding and planning capabilities into a single model without additional training. Leveraging these innovations, our SEA (with only 7B parameters) outperforms existing models of the same parameter scale and achieves performance comparable to larger models (e.g., 32B/72B parameters) on computer use tasks. We plan to release the model weights and related code as open-source resources in the future.

cs.AI↗

arXiv2018

A new centered spatio-temporal autologistic regression model. Application to spatio-temporal analysis of esca disease in a vineyard

We propose a new centered autologistic spatio-temporal model for binary data on a lattice. The centering allows the interpretation of the autoregression coefficients in separating the large scale structure of the model corresponding to an expected mean and the small-scale structure corresponding to the auto-correlation. We discuss the existence of the joint law of the process and show by simulation the interest of this kind of centering. We propose and show the efficiency of the maximum pseudo-likelihood estimator and also a method to choose the best structure of neighborhood. Method is applied to model and fit epidemiological data about Esca disease on a vineyard of the Bordeaux region.

stat.AP↗

arXiv2013

How does degree heterogeneity affect nucleation of Ising model on complex networks?

We investigate the nucleation of Ising model on complex networks and focus on the role played by the heterogeneity of degree distribution on nucleation rate. Using Monte Carlo simulation combined with forward flux sampling, we find that for a weak external field the nucleation rate decreases monotonically as degree heterogeneity increases. Interestingly, for a relatively strong external field the nucleation rate exhibits a nonmonotonic dependence on degree heterogeneity, in which there exists a maximal nucleation rate at an intermediate level of degree heterogeneity. Furthermore, we develop a heterogeneous mean-field theory for evaluating the free-energy barrier of nucleation. The theoretical estimations are qualitatively consistent with the simulation results. Our study suggests that degree heterogeneity plays a nontrivial role in the dynamics of phase transition in networked Ising systems.

cond-mat.stat-mech↗