Unverified50% confidenceFactExact time
DeepSeek-R1-Zero和Qwen系列均大量采用合成数据蒸馏策略
1
Sources
50%
Confidence
Long-term
Relevance
7/18/2026
First Seen
Sources
Related Claims
UnverifiedOpen-source models such as DeepSeek-R1 and Qwen's QwQ are actively exploring a reasoning-enhancement approach similar to o176% similarUnverifiedo1模型的成功启发了DeepSeek-R1、Qwen-QwQ等开源复现工作73% similarUnverifiedDeepSeek-R1 has distilled versions based on knowledge distillation technology.73% similarUnverifiedOpen-source models such as DeepSeek-R1 and Qwen-QwQ are actively exploring slow-thinking inference directions similar to o1.70% similarUnverifiedDeepSeek R1 adopts an open-source strategy, allowing developers to obtain domain-specific models at relatively low computational cost through parameter-efficient fine-tuning techniques such as LoRA and QLoRA.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/550166API
curl https://kongchang.com/api/v1/knowledge/claims/550166MCP
get_claim(id=550166)