DeepSeek-V3 Technical Report
DeepSeek-AI, 2024, DeepSeek-V3 Technical Report (arXiv preprint arXiv:2412.19437)
Subscriber deck — free during early access
1 cards in “5.1 Supervised Fine-Tuning” · whole deck
5.1 Supervised Fine-Tuning
- How did DeepSeek-V3 distill reasoning from DeepSeek-R1 without inheriting its verbosity?Domain expert models were trained with SFT and RL on two kinds of samples: problems with original concise answers, and problems with R1’s long reasoning plus a system prompt asking for reflection and verification. RL taught the experts to blend R1’s patterns into clearer answers; rejection sampling from them then produced the final SFT data.
- DeepSeek-AI, 2024, DeepSeek-V3 Technical Report (arXiv preprint arXiv:2412.19437), §5.1
Source
- DeepSeek-AI. 2024. DeepSeek-V3 Technical Report. In arXiv preprint arXiv:2412.19437. https://arxiv.org/abs/2412.19437