arXiv subjects

Zhuoying Li

Publications and source records attributed to Zhuoying Li.

6 recordsLinked to original sources

arXiv2026

Bernini: Latent Semantic Planning for Video Diffusion

Multimodal large language models (MLLMs) and diffusion models have each reached remarkable maturity: MLLMs excel at reasoning over heterogeneous multimodal inputs with strong semantic grounding, while diffusion models synthesize images and videos with photorealistic fidelity. We argue that these two families can be unified through a simple division of labor: MLLMs perform semantic planning, while diffusion models render pixels from high-level semantic guidance and low-level visual features. Building on this idea, we propose Bernini, a unified framework for video generation and editing. An MLLM-based planner predicts the target semantic representation directly in the ViT embedding space, and a DiT-based renderer synthesizes pixels conditioned on this plan, augmented by text features and, for editing, source VAE features for detail preservation. Because semantics serve as the interface, the planner and renderer can be trained separately and only lightly co-trained, preserving the pretrained strengths of both components while keeping training efficient. To better handle multiple visual inputs, we introduce Segment-Aware 3D Rotary Positional Embedding (SA-3D RoPE), and further incorporate chain-of-thought reasoning in the planner to better transfer understanding into generation. Bernini achieves state-of-the-art performance across a wide range of video generation and editing benchmarks, with the MLLM's pretrained understanding translating into strong generalization on challenging editing tasks.

cs.CV↗

arXiv2025

CSMD: Curated Multimodal Dataset for Chinese Stock Analysis

The stock market is a complex and dynamic system, where it is non-trivial for researchers and practitioners to uncover underlying patterns and forecast stock movements. The existing studies for stock market analysis rely on leveraging various types of information to extract useful factors, which are highly conditional on the quality of the data used. However, the currently available resources are mainly based on the U.S. stock market in English, which is inapplicable to adapt to other countries. To address these issues, we propose CSMD, a multimodal dataset curated specifically for analyzing the Chinese stock market with meticulous processing for validated quality. In addition, we develop a lightweight and user-friendly framework LightQuant for researchers and practitioners with expertise in financial domains. Experimental results on top of our datasets and framework with various backbone models demonstrate their effectiveness compared with using existing datasets. The datasets and code are publicly available at the link: https://github.com/ECNU-CILAB/LightQuant.

cs.CE↗

arXiv2025

Balancing Preservation and Modification: A Region and Semantic Aware Metric for Instruction-Based Image Editing

Instruction-based image editing, which aims to modify the image faithfully according to the instruction while preserving irrelevant content unchanged, has made significant progress. However, there still lacks a comprehensive metric for assessing the editing quality. Existing metrics either require high human evaluation costs, which hinder large-scale evaluation, or are adapted from other tasks and lose task-specific concerns, failing to comprehensively evaluate both instruction-based modification and preservation of irrelevant regions, resulting in biased evaluation. To tackle this, we introduce a new metric called Balancing Preservation and Modification (BPM), tailored for instruction-based image editing by explicitly disentangling the image into editing-relevant and irrelevant regions for specific consideration. We first identify and locate editing-relevant regions, followed by a two-tier process to assess editing quality: Region-Aware Judge evaluates whether the position and size of the edited region align with the instruction, and Semantic-Aware Judge further assesses the instruction content compliance within editing-relevant regions as well as content preservation within irrelevant regions, yielding comprehensive and interpretable quality assessment. Moreover, the editing-relevant region localization in BPM can be integrated into image editing approaches to improve editing quality, demonstrating its broad applicability. We verify the effectiveness of the BPM metric on comprehensive instruction-editing data, and the results show the highest alignment with human evaluation compared to existing metrics, indicating its efficacy. Code is available at: https://joyli-x.github.io/BPM/

cs.GR↗

arXiv2024

Time Series Modeling for Heart Rate Prediction: From ARIMA to Transformers

Cardiovascular disease (CVD) is a leading cause of death globally, necessitating precise forecasting models for monitoring vital signs like heart rate, blood pressure, and ECG. Traditional models, such as ARIMA and Prophet, are limited by their need for manual parameter tuning and challenges in handling noisy, sparse, and highly variable medical data. This study investigates advanced deep learning models, including LSTM, and transformer-based architectures, for predicting heart rate time series from the MIT-BIH Database. Results demonstrate that deep learning models, particularly PatchTST, significantly outperform traditional models across multiple metrics, capturing complex patterns and dependencies more effectively. This research underscores the potential of deep learning to enhance patient monitoring and CVD management, suggesting substantial clinical benefits. Future work should extend these findings to larger, more diverse datasets and real-world clinical applications to further validate and optimize model performance.

cs.LG↗

arXiv2024

AD-Aligning: Emulating Human-like Generalization for Cognitive Domain Adaptation in Deep Learning

Domain adaptation is pivotal for enabling deep learning models to generalize across diverse domains, a task complicated by variations in presentation and cognitive nuances. In this paper, we introduce AD-Aligning, a novel approach that combines adversarial training with source-target domain alignment to enhance generalization capabilities. By pretraining with Coral loss and standard loss, AD-Aligning aligns target domain statistics with those of the pretrained encoder, preserving robustness while accommodating domain shifts. Through extensive experiments on diverse datasets and domain shift scenarios, including noise-induced shifts and cognitive domain adaptation tasks, we demonstrate AD-Aligning's superior performance compared to existing methods such as Deep Coral and ADDA. Our findings highlight AD-Aligning's ability to emulate the nuanced cognitive processes inherent in human perception, making it a promising solution for real-world applications requiring adaptable and robust domain adaptation strategies.

cs.CV↗

arXiv2024

Joint Coverage and Power Control in Highly Dynamic and Massive UAV Networks: An Aggregative Game-theoretic Learning Approach

Unmanned aerial vehicles (UAV) ad-hoc network is a significant contingency plan for communication after a natural disaster, such as typhoon and earthquake. To achieve efficient and rapid networks deployment, we employ noncooperative game theory and amended binary log-linear algorithm (BLLA) seeking for the Nash equilibrium which achieves the optimal network performance. We not only take channel overlap and power control into account but also consider coverage and the complexity of interference. However, extensive UAV game theoretical models show limitations in post-disaster scenarios which require large-scale UAV network deployments. Besides, the highly dynamic post-disaster scenarios cause strategies updating constraint and strategy-deciding error on UAV ad-hoc networks. To handle these problems, we employ aggregative game which could capture and cover those characteristics. Moreover, we propose a novel synchronous payoff-based binary log-linear learning algorithm (SPBLLA) to lessen information exchange and reduce time consumption. Ultimately, the experiments indicate that, under the same strategy-deciding error rate, SPBLLA's learning rate is manifestly faster than that of the revised BLLA. Hence, the new model and algorithm are more suitable and promising for large-scale highly dynamic scenarios.

eess.SY↗