learn.ifs.name

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

DeepSeek-AI, 2025, DeepSeek-R1 (arXiv preprint arXiv:2501.12948)

Subscriber deck — free during early access

1 cards in “2.2.3 Training Template” · whole deck

2.2.3 Training Template

  1. Why is DeepSeek-R1-Zero’s prompt template deliberately minimal?
    It only asks the model to reason inside <think> tags and then give the answer inside <answer> tags, without asking for reflection or any particular strategy, so whatever reasoning behaviors appear come from RL, not from the prompt.
    • DeepSeek-AI, 2025, DeepSeek-R1 (arXiv preprint arXiv:2501.12948), §2.2.3

Source

  • DeepSeek-AI. 2025. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning. In arXiv preprint arXiv:2501.12948. https://arxiv.org/abs/2501.12948