SZ
Sai Qian Zhang
cs.LGcs.AIcs.CVcs.ARcs.DCstat.MLcs.CLComputer Graphicscs.CRcs.GR
On Valency
published · living versionsW_jwa95vfk·v1 · currentpublished
HybridSpec: Throughput-Aware Speculative Decoding for Hybrid GDN-MoE Models
with Haiyu Wang, Zechun Liu, Steven Li
1 version
Preprints & journals
34 papers in the corpus · 2018–2026EgoEverything: A Benchmark for Human Behavior Inspired Long Context Egocentric Video Understanding in AR Environment2604.08342v2 · Qiance Tang, Ziqi Wang, Jieyu Lin et al.2026 · 0 citationsarXiv
pFedNavi: Structure-Aware Personalized Federated Vision-Language Navigation for Embodied AI2602.14401v2 · Qingqian Yang, Hao Wang, Sai Qian Zhang et al.2026 · 0 citationsarXiv
ESCA: Enabling Seamless Codec Avatar Execution through Algorithm and Hardware Co-Optimization for Virtual Reality2510.24787v1 · Mingzhi Zhu, Ding Shang, Sai Qian Zhang2025 · 0 citationsarXiv
QSVD: Efficient Low-rank Approximation for Unified Query-Key-Value Weight Compression in Low-Precision Vision-Language Models2510.16292v1 · Yutong Wang, Haiyu Wang, Sai Qian Zhang2025 · 0 citationsarXiv
Kelle: Co-design KV Caching and eDRAM for Efficient LLM Serving in Edge Computing2510.16040v1 · Tianhua Xia, Sai Qian Zhang2025 · 4 citationsarXiv
CapRecover: A Cross-Modality Feature Inversion Attack Framework on Vision Language Models2507.22828v3 · Kedong Xiu, Sai Qian Zhang2025 · 3 citationsarXiv
A3FR: Agile 3D Gaussian Splatting with Incremental Gaze Tracked Foveated Rendering in Virtual Reality2507.04147v1 · Shuo Xin, Haiyu Wang, Sai Qian Zhang2025 · 3 citationsarXiv
PipeSpec: Breaking Stage Dependencies in Hierarchical LLM Decoding2505.01572v1 · Bradley McDanel, Sai Qian Zhang, Yunhai Hu et al.2025 · 2 citationsarXiv
FovealNet: Advancing AI-Driven Gaze Tracking Solutions for Efficient Foveated Rendering in Virtual Reality.40067704 · Liu, Wenxuan, Duinkharjav, Budmonde, Sun, Qi et al.2025 · 16 citationsIEEE transactions on visualization and computer graphics. 2025;31(5):3183-3193
Foveated Instance Segmentation2503.21854v1 · Hongyi Zeng, Wenxuan Liu, Tianhua Xia et al.2025 · 5 citationsarXiv
Speculative Decoding and Beyond: An In-Depth Survey of Techniques2502.19732v4 · Yunhai Hu, Zining Liu, Zhenyuan Dong et al.2025 · 0 citationsarXiv
HAAN: A Holistic Approach for Accelerating Normalization Operations in Large Language Models2502.11832v1 · Tianfan Peng, Jiajun Qin, Tianhua Xia et al.2025 · 4 citationsarXiv
FovealNet: Advancing AI-Driven Gaze Tracking Solutions for Optimized Foveated Rendering System Performance in Virtual Reality2412.10456v2 · Wenxuan Liu, Monde Duinkharjav, Qi Sun et al.2024 · 0 citationsarXiv
Unlocking Visual Secrets: Inverting Features with Diffusion Priors for Image Reconstruction2412.10448v1 · Sai Qian Zhang, Ziyun Li, Chuan Guo et al.2024 · 1 citationarXiv
GazeGen: Gaze-Driven User Interaction for Visual Content Generation2411.04335v2 · He-Yen Hsieh, Ziyun Li, Sai Qian Zhang et al.2024 · 0 citationsarXiv
PETAH: Parameter Efficient Task Adaptation for Hybrid Transformers in a resource-limited Context2410.17661v2 · Maximilian Augustin, Syed Shakib Sarwar, Mostafa Elhoushi et al.2024 · 0 citationsarXiv
Neural Architecture Search of Hybrid Models for NPU-CIM Heterogeneous AR/VR Devices2410.08326v1 · Yiwei Zhao, Ziyun Li, Win-San Khwa et al.2024 · 0 citationsarXiv
DiTAS: Quantizing Diffusion Transformers via Enhanced Activation Smoothing2409.07756v2 · Zhenyuan Dong, Sai Qian Zhang2024 · 3 citationsarXiv
T3M: Text Guided 3D Human Motion Synthesis from Speech2408.12885v1 · Wenshuo Peng, Kaipeng Zhang, Sai Qian Zhang2024 · 6 citationsarXiv
Rome was Not Built in a Single Step: Hierarchical Prompting for LLM-based Chip Design2407.18276v3 · Andre Nakkab, Sai Qian Zhang, Ramesh Karri et al.2024 · 14 citationsarXiv
DLoRA: Distributed Parameter-Efficient Fine-Tuning Solution for Large Language Model2404.05182v1 · Chao Gao, Sai Qian Zhang2024 · 8 citationsarXiv
BIM: Block-Wise Self-Supervised Learning with Masked Image Modeling2311.17218v1 · Yixuan Luo, Mengye Ren, Sai Qian Zhang2023 · 1 citationarXiv
Hyft: A Reconfigurable Softmax Accelerator with Hybrid Numeric Format for both Training and Inference2311.13290v2 · Tianhua Xia, Sai Qian Zhang2023 · 11 citationsarXiv
CAMEL: Co-Designing AI Models and Embedded DRAMs for Efficient On-Device Learning2305.03148v3 · Sai Qian Zhang, Thierry Tambe, Nestor Cuevas et al.2023 · 1 citationarXiv
SphereFed: Hyperspherical Federated Learning2207.09413v1 · Xin Dong, Sai Qian Zhang, Ang Li et al.2022 · 18 citationsarXiv
A Multi-agent Reinforcement Learning Approach for Efficient Client Selection in Federated Learning2201.02932v1 · Sai Qian Zhang, Jieyu Lin, Qi Zhang2022 · 122 citationsarXiv
FAST: DNN Training Under Variable Precision Block Floating Point with Stochastic Rounding2110.15456v1 · Sai Qian Zhang, Bradley McDanel, H.T. Kung2021 · 72 citationsarXiv
Succinct and Robust Multi-Agent Communication With Temporal Message Control2010.14391v2 · Sai Qian Zhang, Jieyu Lin, Qi Zhang2020 · 39 citationsarXiv
Term Revealing: Furthering Quantization at Run Time on Quantized DNNs2007.06389v2 · H. T. Kung, Bradley McDanel, Sai Qian Zhang2020 · 8 citationsarXiv
On the Robustness of Cooperative Multi-Agent Reinforcement Learning2003.03722v1 · Jieyu Lin, Kristina Dzeparoska, Sai Qian Zhang et al.2020 · 66 citationsarXiv
RTN: Reparameterized Ternary Network1912.02057v2 · Yuhang Li, Xin Dong, Sai Qian Zhang et al.2019 · 28 citationsarXiv
Efficient Communication in Multi-Agent Reinforcement Learning via Variance Based Control1909.02682v2 · Sai Qian Zhang, Qi Zhang, Jieyu Lin2019 · 60 citationsarXiv
Full-stack Optimization for Accelerating CNNs with FPGA Validation1905.00462v1 · Bradley McDanel, Sai Qian Zhang, H. T. Kung et al.2019 · 1 citationarXiv
Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations: Column Combining Under Joint Optimization1811.04770v1 · H. T. Kung, Bradley McDanel, Sai Qian Zhang2018 · 22 citationsarXiv
Career total: 79 works. 34 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.