ChenxinAn-fdu/POLARIS

post-training recipe using RL to boost reasoning in language models

View on index · View in 3D Map

Related repos

• parallel training for language models with dataset prep and programmatic access
• open-source reinforcement learning for language models
• open notes and guides for training large language models
• Executes reasoning paths in parallel with language models
• large language model tutorials and training basics from scratch
• reinforcement learning library for training large language models, supports many RL algorithms, integrates with LLM tools, scales to models with up to 671B parameters
• open-source large language model focused on complex reasoning
• Graph-R1 lets language models reason over knowledge graphs using multi-turn interactions and reinforcement learning for better question answering
• Step-by-step code for adding reasoning to a pre-trained LLM, with hands-on examples and explanations
• Reinforcement learning framework for language agents