Qizhe Zhang

FasterVLM. Official code for paper: [CLS] Attention is All You Need for Training-Free Visual Token Pruning: Make VLM Inference Faster.

CDPruner. [NeurIPS 2025] Official code for paper: Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs.

VisPruner. [ICCV 2025] Official code for paper: Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs

MahjongZero. MahjongZero: DouZero for Mahjong | 麻将AI

BiCross-old. BiCross (Unsupervised Spike Depth Estimation via Cross-modality Cross-domain Knowledge Transfer).

MoSA. Official code of MoSA (Mixture of Sparse Adapters).

PKUElective2022Spring. PKUAutoElective for 2022 spring semester

semantic-poss-api. SemanticPOSS API for visualizing dataset, processing data, and evaluating results.

BiCross. Official code of BiCross (Unsupervised Spike Depth Estimation via Cross-modality Cross-domain Knowledge Transfer) [ICRA 2024].

EMA-and-Ensemble-Lip-Networks. Implementation of Boosting Certified $\ell_\infty$-dist Robustness with EMA Method and Ensemble Model

DAIR-V2X. A modified version of the OpenDAIRV2X for the competition of VIC3D Object Detection challenge.

SparseVLM. Python

MedSegX. Official code for A generalist foundation model and database for open-world medical image segmentation (MedSegX).

Render. Computer Generated Imagery Project3 - Render. Scripts are written in C++ and compiled by xmake.

U-RISC. EM cytomembrane segmentation with neural network ensemble and morphological processing.

Interactive-Character-Animation. Computer Generated Imagery Project1 - Interactive Character Animation. Scripts are written in C# and rendered in Unity.

Railway-Ticket-Identification. DIP 2021 Fall Final Project - Railway Ticket Identification

Interactive-Multiple-Views. Interactive Multiple Views Visualization of Ming Jinshi List from CBDB.

VET. Visual Efficient Tuning. (PEFT v.s. FT)