tracked author
Xingkai Yu (俞星凯)
DeepSeekMoE, DeepSeek-R1, Engram and DSpark recurring author. GitHub profile identifies Xingkai Yu and pins DeepSeek-V3, DeepSeek-R1 and nano-vllm; Scholar records DeepSeek AI publications. No high-confidence personal X account was found.
Related Notes
按论文归档时间排序，展示该作者在本站已经出现的材料。
DSpark: Confidence Scheduled Speculative Decoding with Semi Autoregressive Generation
DSpark 是 DeepSeek 把并行 drafter 推向生产 serving 的一套完整方案：用 DFlash 式 parallel backbone 先一次生成长候选块，再用低秩 Markov head 注入块内局部自回归依赖，随后用 calibrated confidence head 和硬件感知 prefix scheduler 按请求与负载动态裁剪 target verification 长度。离线 Qwen3 / G...
DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models
DeepSeekMoE 把 MoE (Mixture of Experts，混合专家) 的效率问题改写成专家专门化问题：fine grained expert segmentation 通过切小 FFN (Feed Forward Network，前馈网络) 专家并增加激活专家数，提高每个 token 的专家组合分辨率；shared expert isolation 通过固定激活共享专家承载通用知识，让 routed experts ...
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
Engram 把大模型稀疏性从 MoE (Mixture of Experts，混合专家，用条件计算扩大 total 参数并控制 active compute) 扩展到 conditional memory：用 hashed N gram lookup 存静态局部模式，用 context aware gate 决定是否把查到的记忆注入 hidden state；在 iso parameter / iso FLOPs 的 27B MoE...
DeepSeek R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
DeepSeek R1 v2 的核心结论是：大规模 outcome based RL 可以在强 base model 上诱导 long CoT reasoning、自我反思、验证和策略切换等行为；R1 Zero 证明无需 SFT 也能通过 rule based verifiable reward 激发 reasoning capability，R1 则通过 cold start SFT、两阶段 RL、rejection samplin...