[NeurIPS'25 Spotlight] (DANCE) Disentangled Concepts Speak Louder Than Words:Explainable Video Action Recognition official code repository

LeetGPU Challenges

A reading list for large models safety, security, and privacy (including Awesome LLM Security, Safety, etc.).

Accepted by ECCV 2024

Awesome Unified Multimodal Models

[ICML 2025] Unlearning in Diffusion Models using Sparse Autoencoders

Lightweight nudity detection

《大语言模型》作者：赵鑫，李军毅，周昆，唐天一，文继荣

[NeurIPS 2025] Official implementation of "XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation".

[SIGGRAPH Asia 2025] DreamO: A Unified Framework for Image Customization

Config files for my GitHub profile.

OI / ACM-ICPC essays and learning materials

This repository is the homework about Intelligent Science and Technology in NJUST（南京理工大学 智能科学与技术 作业汇总）

简易的yolo半自动标注库，目前只支持单目标。如果数据集图片背景复杂，可能工作量不比直接标的小，因为sam是通用的分割模型。但是可以适当通过调整参数修改。

My Blog (https://www.zhangwp.com).

贵校课程资料民间整理