This is a heavily interactive web application, and JavaScript is required. Simple HTML interfaces are possible, but that is not what this is.
Post
Gradient Flow
gradientflow.com
did:plc:ahtyudnsxy3he5ynxbijws6r
Researchers from China explore the Secret Behind OpenAI o1's Success
- o1 marks shift toward RL
- and a shift from scaling solely training computation to scaling both training & inference
- o1 architecture shows how combining learning & search drives advancement in LLMs
arxiv.org/abs/2412.14135
https://arxiv.org/abs/2412.14135
2024-12-29T15:40:16.149Z