Minimal reproduction of DeepSeek R1-Zero

DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Train transformer language models with reinforcement learning.

📚 从零开始的智能体原理与实践教程

Democratizing Reinforcement Learning for LLMs

Repository for research works and resources related to model reprogramming <https://arxiv.org/abs/2202.10629>

AgentFlow: In-the-Flow Agentic System Optimization

Code and dataset for paper: DeepPlanner: Scaling Planning Capability for Deep Research Agents via Advantage Shaping

[NeurIPS 2025 Spotlight] "Physics-Driven Spatiotemporal Modeling for AI-Generated Video Detection"

Official implementation of Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation

[ICLR2021 Oral] Free Lunch for Few-Shot Learning: Distribution Calibration

Official code for "Vision Transformers with Self-Distilled Registers" (NeurIPS 2025 Spotlight)

source code for NeurIPS'24 paper "HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detection"

Learning Deep Representations of Data Distributions

Qwen Code is a coding agent that lives in the digital world.

从无名小卒到大模型（LLM）大英雄~ 欢迎关注后续！！！

This is the homepage of a new book entitled "Mathematical Foundations of Reinforcement Learning."

🦜🔗 The platform for reliable agents.

Qwen-Image is a powerful image generation foundation model capable of complex text rendering and precise image editing.

scikit-learn cross validators for iterative stratification of multilabel data

Pytorch implementation of Learning in Imperfect Environment: Multi-Label Classification with Long-Tailed Distribution and Partial Labels

Empirical tricks for training robust models (ICLR 2021)

Home of StarCoder2!

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (V…

This may be the simplest implement of DDPM. You can directly run Main.py to train the UNet on CIFAR-10 dataset and see the amazing process of denoising.

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.

ArcherCodeR is an open-source initiative enhancing code reasoning in large language models through scalable, rule-governed reinforcement learning.

[arXiv:2508.00410] "Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models"

[arXiv:2508.00410] "Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models"