Kimi's MoE Architecture Traced Back to Yang Zhilin's Decade-Old PhD Thesis
Following Yang Zhilin's GTC 2026 keynote, tech communities are tracing Kimi's competitive MoE architecture back to his decade-old PhD research on XLNet at CMU. His early academic work laid the crucial groundwork for the current trillion-parameter model.
2026-07-21 ~ 2026-07-21 · 3 related posts
- Kimi’s MoE lineage may trace back to Yang Zhilin’s decade-old XLNet PhD work — FinanceYF5 · 2026-07-21
- Kimi's Edge: Tracing Founder Yang Zhilin's Decade-Old PhD Thesis — FinanceYF5 · 2026-07-21
- Yang Zhiling’s 10-year-old PhD work may have shaped Kimi K2’s trillion-parameter MoE — FinanceYF5 · 2026-07-21