6 papers
Decoupled Residual Denoising Diffusion Models for Unified and Data Efficient Image-to-Image Translation
Ziyue Lin, Jiahe Hou, Hongyu Xia +6
We propose Decoupled Residual Denoising Diffusion models (DRDD) for unified and data-efficient image-to-image (I2I) translation. While diffusion models have advanced I2I translatio…
ACE-Sync: An Adaptive Cloud-Edge Synchronization Framework for Communication-Efficient Large-Scale Distributed Model Training
Yi Yang, Ziyu Lin, Liesheng Wei
Large-scale deep learning models impose substantial communication overh ead in distributed training, particularly in bandwidth-constrained or heterogeneous clo ud-edge environments…
Unleashing the Potential of Large Language Models for Text-to-Image Generation through Autoregressive Representation Alignment
Xing Xie, Jiawei Liu, Ziyue Lin +4
We present Autoregressive Representation Alignment (ARRA), a new training framework that unlocks global-coherent text-to-image generation in autoregressive LLMs without architectur…
A Cloud-Based Spatio-Temporal GNN-Transformer Hybrid Model for Traffic Flow Forecasting with External Feature Integration
Zhuo Zheng, Lingran Meng, Ziyu Lin
Accurate traffic flow forecasting is essential for the development of intelligent transportation systems (ITS), supporting tasks such as traffic signal optimization, congestion man…
Design and Implementation of Code Completion System Based on LLM and CodeBERT Hybrid Subsystem
Bingbing Zhang, Ziyu Lin, Yingxin Su
In the rapidly evolving industry of software development, coding efficiency and accuracy play significant roles in delivering high-quality software. Various code suggestion and com…
FedVLMBench: Benchmarking Federated Fine-Tuning of Vision-Language Models
Weiying Zheng, Ziyue Lin, Pengxin Guo +3
Vision-Language Models (VLMs) have demonstrated remarkable capabilities in cross-modal understanding and generation by integrating visual and textual information. While instruction…