- All languages
- Assembly
- AutoHotkey
- B4X
- C
- C#
- C++
- CMake
- CSS
- Cuda
- Dart
- Dockerfile
- Fancy
- Fluent
- Go
- HTML
- Haskell
- Java
- JavaScript
- Jinja
- Jupyter Notebook
- Kotlin
- Lua
- MATLAB
- MDX
- Makefile
- Markdown
- Mustache
- Objective-C
- PHP
- Perl
- Python
- Rich Text Format
- Roff
- Ruby
- Rust
- SCSS
- Scala
- Shell
- Svelte
- Swift
- TeX
- TypeScript
- Visual Basic 6.0
- Vue
- XSLT
- Zig
Starred repositories
Covers pre-training data, Tokenizer, Transformer, MoE,distributed training, Scaling Laws, inference & alignment .6 progressive code assignments for full-stack LLM learning | 涵盖预训练数据、分词器、Transformer…
[ACL'26 Oral] AgentOCR is a token-efficient framework that compresses multi-turn agent history by rendering it into images and adopting RL-driven self-compression
Type-safe, extensible, and high-performance background processing library for Rust
Eagle: Frontier Vision-Language Models with Data-Centric Strategies
A full model customization repo for Liquid Foundation Models
BoT-SORT: Robust Associations Multi-Pedestrian Tracking
RTMPose series (RTMPose, DWPose, RTMO, RTMW) without mmcv, mmpose, mmdet etc.
[ECCV 2022] ByteTrack: Multi-Object Tracking by Associating Every Detection Box
A Chinese probabilistic decision model: one forward pass, straight to candidate probabilities
[AAAI2026] X-SAM: From Segment Anything to Any Segmentation
WiLoR: End-to-end 3D hand localization and reconstruction in-the-wild
Includes the code for training and testing the CountGD model from the paper CountGD: Multi-Modal Open-World Counting.
Official PyTorch implementation of "Multi-modal Queried Object Detection in the Wild" (accepted by NeurIPS 2023)
Code for OPUS: A Simple yet Effective Unified Framework for Open-Vocabulary Detection
Multimodal Reference Visual Grounding (will have more updates)
[IROS 2025] NIDS-Net: A unified framework for novel instance detection and segmentation
SOTA Re-identification Methods and Toolbox
1M param OCR model - runs on CPU, 15x faster than tesseract, close to 1B+ param models in accuracy
Document image dewarping library using a cubic sheet model
Improving and Evaluating Hand-Object Interaction Detection
VideoJev: System One Model for Video Understanding and Interaction
对招股书的财务勾稽关系校验,包含跨表勾稽、表内勾稽、文表勾稽多种校验,校验后在pdf文档中自动高亮显示。
2022阿里天池真实场景篡改图像检测挑战赛-冠军方案(1/1149)
GGUF quantizations of Xiaomi-OCR-0 (0.8B document-parsing VLM) for llama.cpp, with reproducible scripts and measured accuracy/speed/memory