new blog post "There Are No New Ideas In AI.... Only New Datasets" in which i summarize LLMs in exactly four breakthroughs and explain why it was really *data* all along that mattered... not algorithms
62
306
44
2,796
Replying to @jxmnop
RLHF was introduced earlier in the 2017 paper "Deep Reinforcement Learning from Human Preferences" by Paul Christiano

Apr 10, 2025 · 6:24 AM UTC
2