This is a heavily interactive web application, and JavaScript is required. Simple HTML interfaces are possible, but that is not what this is.
Post
Seth Karten
sethkarten.ai
did:plc:rk7mvfnpoch5c5awt4tqaoro
Is GRPO sufficient? Let's look at the OG AlphaGo paper -->
Policy network: yes, from LLM
Rollouts: yes, from advantage calculation
Value network: no
Perhaps we are in the REINFORCE era of RL for LLMs still
2025-02-13T20:31:10.153Z