burny pushes back on 'LLMs just predict likely text': RLHF, CoT and tools changed that
burny_tech · x · 2026-10-01
Leaker burnytech quotes tech journalist Taylor Lorenz and argues her framing is outdated: the 'LLMs just output the most likely text sequences' mental model is 2+ years old. Modern models' output distributions are heavily shaped by RL from rewards, plus CoT, tool use and RAG. Lorenz agrees frontier models are great but hopes error rates improve.
More from Models
- Speculation: rumored Gemini 4 'Argon' tier looks like Google's Sonnet-class rival — teortaxesTex · 2026-10-01
- Light user burns through entire $200/month Codex plan in a few tasks — barney · 2026-10-01
- Sol 6.1 called a strong answer to Opus 5.5, arguably beating Astra in some ways — teortaxesTex · 2026-10-01
- No, open vs closed model usage didn't flip 80:20 in 12 weeks — analyst fact-checks viral claim — AccBalanced · 2026-10-01
- OpenAI's $300 Ultrafast mode hits 70tps while rivals match it at $20-$50 — NandaVegg · 2026-10-01
- First look at Opus 5.5 building an SCP-096 game in one go — imjustnewatai · 2026-10-01