Author

Zhendong Mi

Observed name: Mi, Zhendong

Institution: Stevens Institute of Technology

Observed institution: Stevens Institute of Technology

ORCID: 0009-0008-2907-9839

Observed ORCID: 0009-0008-2907-9839

OpenAlex ID: A5105697701

Effective MoE-based LLM Compression by Exploiting Heterogeneous Inter-Group Experts Routing Frequency and Information Density

2026 · Mi, Zhendong, Chen, Yixiao, et al. · arXiv (Cornell University) · 8

KerZOO: Kernel Function Informed Zeroth-Order Optimization for Accurate and Accelerated LLM Fine-Tuning

2025 · Mi, Zhendong, Tan, Qitao, et al. · arXiv (Cornell University) · 5

ACE: Exploring Activation Cosine Similarity and Variance for Accurate and Calibration-Efficient LLM Pruning

2025 · Mi, Zhendong, Kong, Zhenglun, et al. · arXiv (Cornell University) · 4

Towards Fast LLM Fine-tuning through Zeroth-Order Optimization with Projected Gradient-Aligned Perturbations

2025 · Mi, Zhendong, Tan, Qitao, et al. · arXiv (Cornell University) · 3

Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression

2026 · Zhu, Hengyi, Mi, Zhendong, et al. · arXiv (Cornell University) · 1

Characterizing Performance Tradeoffs in Heterogeneous GPU-Wafer-Scale Accelerator Systems

2026 · Hu, Ziyu, Ye, Zhijing, et al. · arXiv (Cornell University)

SpectralCache: Accelerating Diffusion-Based World Models via Spectral Feature Caching

2026 · Mi, Zhendong, Zhao, Pu, et al. · arXiv (Cornell University)

Imagine the Future, Internalize the Gist: Efficient VLA Reasoning via Internalized Spatiotemporal Imagination

2026 · Li, Shenglan, Mi, Zhendong, et al. · arXiv (Cornell University)

Sampling via Decision-Flow: Training-Free Extraction of Improved Latent Reasoning Paths in Large Language Models

2026 · Mi, Zhendong, Huang, Shaoyi · arXiv (Cornell University)

Late Breaking Results: Structured Expert Routing for Efficient MoE Inference under GPU-CPU Orchestration

2026 · Yixiao Chen, Arman Akbari, et al. · ACM/IEEE Design Automation Conference (DAC)

Works count: 12

Citation count: 48

H-index: 4

I10-index: 2