tracked author

Yang Xu (徐旸)

MiniMax Sparse Attention coauthor. Homepage identifies him as a Zhejiang University undergraduate and MiniMax intern working on LLM pretraining and model architecture; GitHub links the same homepage, Scholar and X handle. This Yang Xu is kept separate from other same-name authors unless future evidence proves identity overlap.

2 archived notes X: high HomepageGitHubX

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。

Jun 24, 2026

MiniMax Sparse Attention

MiniMax Sparse Attention (MSA) 是 MiniMax M3 的长上下文技术核心:它在 GQA (Grouped Query Attention,分组查询注意力) 之上增加轻量 Index Branch,为每个 query token 和每个 GQA group 选择少量 KV blocks,再由 Main Branch 对被选 block 做精确 softmax attention;论文的强证据在于 109...

2606.13392-minimax-sparse-attention-m3 Sparse AttentionLong ContextMoE Architecture
Jun 10, 2026

Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection Sampling

Bebop 的核心判断是:MTP 在 RL rollout 中失速的主因主要来自 policy entropy fluctuation,frozen MTP head 与更新后 policy 的权重漂移影响较小;因此有效方案是把 acceptance method 换成 probabilistic rejection sampling,并用 end to end TV loss 在 RL 前训练 MTP heads,使 draft t...

2606.12370-bebop-mtp-rejection-sampling-rl-training Rollout OptimizationSpeculative DecodingMulti-Token Prediction