tracked author
Yu Wu (吴俣)
Math-Shepherd coauthor and DeepSeek LLM Alignment Team lead. Homepage states that his team pioneered GRPO and DeepSeek-R1-Zero; Google Scholar identifies him as Yu Wu (吴俣), DeepSeek AI. X search result matches the same name and DeepSeek technical-staff role.
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 ...
DeepSeek V3 Technical Report
DeepSeek V3 用 MLA (Multi head Latent Attention，多头潜变量注意力)、DeepSeekMoE、无辅助损失负载均衡、MTP (Multi Token Prediction，多 token 预测)、FP8 混合精度训练和 DualPipe 通信重叠，把 671B total / 37B active 的开放 MoE (Mixture of Experts，混合专家) 模型训练到强代码、数学和通用...
Math Shepherd: Verify and Reinforce LLMs Step by step without Human Annotations
Math Shepherd 的核心价值在于把数学推理步骤的标注问题改写为“当前 step 之后还能否补全到正确答案”的 Monte Carlo potential estimation：对每个中间 step 采样多个 continuation，用最终答案正确性给 step 生成 hard / soft pseudo label，训练 PRM 做 verifier reranking，并进一步把 PRM reward 接入 step b...
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...