# @davidad — 2025-02-11

♥374 ↻31 · https://x.com/davidad/status/1889278316005208387

I never saw this snippet of the DeepSeek-R1-Zero paper on my timeline, so many of you may not have seen it yet.Basically, informal backtracking search becomes emergent from RL when the base LLM is big enough.The authors are clearly astonished by this, and you should be too. https://t.co/PhUgyL8K6c

tags: author:davidad, kind:tweet, model:deepseek-r1, model:deepseek-r1-zero, on:deepseek-r1, on:deepseek-r1-zero, year:2025
cited on: _dossiers/deepseek-r1-zero.md, _dossiers/deepseek-r1.md, deepseek-r1, deepseek-r1-zero