Contexts Optical Compression

Display giant ASCII-art logos with colorful gradients in your terminal — like Claude Code or Gemini CLI.

[Preprint 2025] Ditto: Scaling Instruction-Based Video Editing with a High-Quality Synthetic Dataset

Thinking with Videos from Open-Source Priors. We reproduce chain-of-frames visual reasoning by fine-tuning open-source video models. Give it a star 🌟 if you find it useful.

The offical repository of "So-Fake: Benchmarking and Explaining Social Media Image Forgery Detection"

[NeurIPS 2025 DB] OneIG-Bench is a meticulously designed comprehensive benchmark framework for fine-grained evaluation of T2I models across multiple dimensions, including subject-element alignment,…

NEO Series: Native Vision-Language Models from First Principles

Official implementation of "UniLiP: Adapting CLIP for Unified Multimodal Understanding, Generation and Editing"

本项目是博主在知识星球编写的文章项目，包含了多个项目的部署教程

HALO: Hadamard-Assisted Low-Precision Optimization and Training method for finetuning LLMs. 🚀 The official implementation of https://arxiv.org/abs/2501.02625

This is the repository for the paper ‘A Survey of Inductive Reasoning for Large Language Models’

The best ChatGPT that $100 can buy.

OpenVLA: An open-source vision-language-action model for robotic manipulation.

Tora: Torchtune-LoRA for RL

The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

VChain: Chain-of-Visual-Thought for Reasoning in Video Generation

VideoNSA: Native Sparse Attention Scales Video Understanding

Automatic Video Generation from Scientific Papers

大麦自动抢票，支持人员、城市、日期场次、价格选择

An Industrial-Level Controllable and Efficient Zero-Shot Text-To-Speech System

Code release for Ming-UniVision: Joint Image Understanding and Geneation with a Continuous Unified Tokenizer

A curated list of resources on Reinforcement Learning with Verifiable Rewards (RLVR) and the reasoning capability boundary of Large Language Models (LLMs).

[NeurIPS 2025] TTRL: Test-Time Reinforcement Learning

[NeurIPS'25] DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution

[NeurIPS 2025] Inference-Time Text-to-Video Alignment with Diffusion Latent Beam Search

An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

📄 Awesome CV is LaTeX template for your outstanding job application

Kandinsky 5.0: A family of diffusion models for Video & Image generation