PrototypeCommon
McGill-NLP/VinePPO
Code for the paper "VinePPO: Unlocking RL Potential For LLM Reasoning Through Refined Credit Assignment"
- Owner
- McGill-NLP (organization)
- Stars
- Stars: 197
- Forks
- Forks: 23
- Language
- Python
- License
- MIT
- Created
- 2024
- Tier
- Prototype (100 to 249 ★) What this tier means
More repos by McGill-NLP
All 9 repos by McGill-NLP- McGill-NLP/llm2vecStars: 1,729PythonBeta
- McGill-NLP/webllamaStars: 1,398PythonBeta
- McGill-NLP/nano-aha-momentStars: 640Jupyter NotebookAlpha
- McGill-NLP/the-markovian-thinkerStars: 351PythonAlpha
- McGill-NLP/medalStars: 285PythonAlpha
- McGill-NLP/weblinxStars: 165PythonPrototype
Nearby in Python
All Python repos- Davipar/djev-devStars: 197PythonPrototype
- kusaida/AdwairStars: 197PythonPrototype
- xxPcy/PharmaLLM-AgentStars: 197PythonPrototype
- llimllib/pymag-treesStars: 197PythonPrototype
- msoulier/tftpyStars: 197PythonPrototype
- andreisavu/mongodb-logStars: 197PythonPrototype