Yuxuan Li

minit2i-torch. Official PyTorch re-implementation of MiniT2I.

minimind-o. 🎙️ 「大模型」从0训练0.1B能听能说能看的全模态Omni模型！A 0.1B Omni model trained from scratch, capable of listening, speaking, and seeing!

academic-research-skills. Academic Research Skills for Claude Code: research → write → review → revise → finalize

Embodied-AI-Guide. [Lumina具身智能社区] 具身智能技术指南 Embodied-AI-Guide

andrej-karpathy-skills. A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.

nkuthesis. Nankai University Thesis LaTeX Template (南开大学硕博毕业论文模版 v2026.5)

WOW-Seg-Meta. Python

page-agent. JavaScript in-page GUI agent. Control web interfaces with natural language.

crystal. official repository of article "CrystaL: Spontaneous Emergence of Visual Latents in MLLMs"

ASID-Caption. ASID-Caption: Attribute-Structured and Quality-Verified Audiovisual Instruction Dataset and Training Pipeline for Fine-Grained Video Understanding.

stagehand. The SDK For Browser Agents

openclaw. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

visionary. Visionary: The World Model Carrier Built on WebGPU-Powered Gaussian Splatting Platform

TrendRadar. ⭐AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts.🎯 告别信息过载，你的 AI 舆情监控助手与热点筛选工具！聚合多平台热点 + RSS 订阅，支持关键词精准筛选。AI 智能筛选新闻 + AI 翻译 + AI 分析简报直推手机，也支持接入 MCP 架构，赋能 AI 自然语言对话分析、情感洞察与趋势预测等。支持 Docker ，数据本地/云端自持。集成微信/飞书/钉钉/Telegram/邮件/ntfy/bark/slack 等渠道智能推送。

RSCoVLM. [Remote Sensing 2026] Co-Training Vision Language Models for Remote Sensing Multi-task Learning

JiT. PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

DeepOCR. A reproduction of the Deepseek-OCR model including training

ViTP. Offical implementation of "Visual Instruction Pretraining for Domain-Specific Foundation Models"

Rex-Omni. [CVPR2026] Detect Anything via Next Point Prediction

SE-GUI. [NeurIPS 2025]"Enhancing Visual Grounding for GUI Agents via Self-Evolutionary Reinforcement Learning"

flagged_pointer. Rust

dinov3. Reference PyTorch implementation and models for DINOv3

pix2seq. Pix2Seq codebase: multi-tasks with generative modeling (autoregressive and diffusion)

ai4rs. AI for remote sensing, remote sense, object detection, oriented object detection, computer vision, cv

MODA. [ICML 2025 Spotlight] MODA: MOdular Duplex Attention for Multimodal Perception, Cognition, and Emotion Understanding

REG. [NeurIPS 2025 Oral] Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Think

LLaVA-Scissor. The official code for the paper: LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMs

one-shot-em. One-shot Entropy Minimization

MMaDA. MMaDA - Open-Sourced Multimodal Large Diffusion Language Models (dLLMs with block diffusion, mixed-CoT, unified RL)

Sa2VA. Official Repo For Pixel-LLM Codebase: Sa2VA (Arxiv-25), SAMTok (CVPR-26), VRT, SaSaSa2VA (1-st solution for LSVOS)

SARLANG-1M. Official PyTorch Implementation of SARLANG-1M: A Benchmark for Vision-Language Modeling in SAR Image Understanding [IEEE TGRS 2026].

RISEBench. [NIPS 2025 DB Oral] Official Repository of paper: Envisioning Beyond the Pixels: Benchmarking Reasoning-Informed Visual Editing

InternVL-MMDetSeg. Train InternViT-6B in MMSegmentation and MMDetection with DeepSpeed

Mr.DETR. [CVPR 2025] Mr. DETR: Instructive Multi-Route Training for Detection Transformers

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

LlamaFactory. Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Visual-RFT. Official repository of 'Visual-RFT: Visual Reinforcement Fine-Tuning' & 'Visual-ARFT: Visual Agentic Reinforcement Fine-Tuning'’

open-r1. Fully open reproduction of DeepSeek-R1

SARChat. The first large-scale multimodal dialogue dataset focusing on Synthetic Aperture Radar (SAR) imagery.

VLM-R1. Solve Visual Understanding with Reinforced VLMs

MoE-LLaVA. 【TMM 2025🔥】 Mixture-of-Experts for Large Vision-Language Models

Visual-CoT. [Neurips'24 Spotlight] Visual CoT: Advancing Multi-Modal Language Models with a Comprehensive Dataset and Benchmark for Chain-of-Thought Reasoning

LISA. Project Page for "LISA: Reasoning Segmentation via Large Language Model"

segllm. Code release for "SegLLM: Multi-round Reasoning Segmentation"

sood-mcl. [AAAI2025] Multi-clue Consistency Learning to Bridge Gaps Between General and Oriented Object in Semi-supervised Detection

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

mllm-mmrotate. [IGARSS 2025 Oral] A Simple Aerial Detection Baseline of Multimodal Language Models.

Text4Seg. [ICLR2025] Text4Seg: Reimagining Image Segmentation as Text Generation

nk-remote. Python

techniques. Techniques for deep learning with satellite & aerial imagery

Strip-R-CNN. [AAAI 2026]"Strip R-CNN: Large Strip Convolution for Remote Sensing Object Detection"

learning_research. 本人的科研经验

DeepSeek-V3. Python

RSAR. [CVPR 2025] Official implementation for the paper "RSAR: Restricted State Angle Resolver and Rotated SAR Benchmark".

RSUniVLM. Python

VHM. VHM: Versatile and Honest Vision Language Model for Remote Sensing Image Analysis

Awesome-Remote-Sensing-Multimodal-Large-Language-Model. Multimodal Large Language Models for Remote Sensing (RS-MLLMs): A Survey

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

AI-TOD-R. Python

InternVL. [CVPR 2024 Oral] InternVL Family: A Pioneering Open-Source Alternative to GPT-4o. 接近GPT-4o表现的开源多模态对话模型

GeoChat. [CVPR 2024 🔥] GeoChat, the first grounded Large Vision Language Model for Remote Sensing

EarthGPT.

DenseVLM. [ICCV 2025] Unbiased Region-Language Alignment for Open-Vocabulary Dense Prediction

lyhisme.github.io. AcadHomepage: A Modern and Responsive Academic Personal Homepage

mvits_for_class_agnostic_od. [ECCV'22] Official repository of paper titled "Class-agnostic Object Detection with Multi-modal Transformer".

ultralytics. Ultralytics YOLO26, YOLO11, YOLOv8 — object detection, instance segmentation, semantic segmentation, image classification, pose estimation, object tracking

HazyDet. Python

GrokLST. Python

FreqFusion. TPAMI：Frequency-aware Feature Fusion for Dense Image Prediction

MixPL. Mixed Pseudo Labels for Semi-Supervised Object Detection

AlignDet. [ICCV 2023] AlignDet: Aligning Pre-training and Fine-tuning in Object Detection.

ConsistentTeacher. [CVPR2023 Highlight] Consistent-Teacher: Towards Reducing Inconsistent Pseudo-targets in Semi-supervised Object Detection

InstanceLoc. [CVPR 2021] Instance Localization for Self-supervised Detection Pretraining

SlotCon. (NeurIPS 2022) Self-Supervised Visual Representation Learning with Semantic Grouping

SoCo. [NeurIPS 2021 Spotlight] Aligning Pretraining for Detection via Object-Level Contrastive Learning

nxtp. PyTorch Implementation of Object Recognition as Next Token Prediction [CVPR'24 Highlight]

materials_discovery. Jupyter Notebook

RSUL. A large remote sensing unlabeled dataset for semi-supervised oriented object detection.

NAIP. This repository contains the official implementation for the series of NAIP family.

open-cd. A Change Detection Repo Standing on the Shoulders of Giants

DroneVehicle. Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning

Achievement-based-MTL. Source Code of the paper "Achievement based Training Progress Balancing for Multi-Task Learning" accepted in ICCV2023

AutomaticWeightedLoss. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics, Auxiliary Tasks in Multi-task Learning

UniDetector. Code release for our CVPR 2023 paper "Detecting Everything in the Open World: Towards Universal Object Detection".

GrokSAR. Python

balatro-bench. TypeScript

nk-det. An open source codebase for object detection based on Jittor

OmniCorpus. [ICLR 2025 Spotlight] OmniCorpus: A Unified Multimodal Corpus of 10 Billion-Level Images Interleaved with Text

STAR-MMRotate. [TPAMI] Oriented object detection on STAR dataset.

Cascade-CLIP. Official implement of ICML2024 Cascade-CLIP: Cascaded Vision-Language Embeddings Alignment for Zero-Shot Semantic Segmentation

PIIP. [NeurIPS 2024 Spotlight ⭐️ & TPAMI 2025] Parameter-Inverted Image Pyramid Networks (PIIP)

UniDet. Object detection on multiple datasets with an automatically learned unified label space.

MiDaS. Code for robust monocular depth estimation described in "Ranftl et. al., Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer, TPAMI 2022"

open_clip. An open source implementation of CLIP.

DeepSeek-Coder. DeepSeek Coder: Let the Code Write Itself

ETSN. Efficient Two-Step Networks for Temporal Action Segmentation (Neurocomputing 2021)

DeST. An official code for "A Decoupled Spatio-Temporal Framework for Skeleton-based Action Segmentation".

Paraformer. Paraformer: Updating Large-scale High-resolution Land Cover Maps from Low-resolution Historical Labels

TR-DETR. Official pytorch repository for "TR-DETR: Task-Reciprocal Transformer for Joint Moment Retrieval and Highlight Detection" (AAAI 2024 Paper)

M2PT. [CVPR 2024] Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities