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
UnverifiedDeepSeek-R1、Qwen的QwQ等开源模型在积极探索类似o1的推理增强路线76% similarUnverifiedo1模型的成功启发了DeepSeek-R1、Qwen-QwQ等开源复现工作73% similarUnverifiedDeepSeek-R1有基于知识蒸馏技术的蒸馏版本73% similarUnverifiedDeepSeek-R1和Qwen-QwQ等开源模型在积极探索类似o1的慢思考推理方向70% similarUnverifiedDeepSeek R1采取开源策略，开发者可通过LoRA、QLoRA等参数高效微调技术以较低算力成本获得垂直领域专属模型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)