Kimi’s MoE lineage may trace back to Yang Zhilin’s decade-old XLNet PhD work
FinanceYF5 · x · 2026-07-21
The post argues that one key technical source behind Kimi’s success may be founder Yang Zhilin’s PhD work from a decade ago.
Background
- Yang is 34, studied at Tsinghua and CMU, and spent his PhD years at Meta AI and Google Brain.
- His widely cited XLNet paper has reportedly been cited more than 10,000 times.
Claim
- The poster suggests the ideas behind XLNet later evolved into the trillion-parameter MoE architecture used by Kimi K2.
- In other words, the model’s architecture lineage may trace back to a long research path rather than a sudden product breakthrough.
Related event: Kimi's MoE Architecture Traced Back to Yang Zhilin's Decade-Old PhD Thesis(3 posts)→
More from Companies & People
- Forter's 13 lessons from its agent sprint: skip custom RAG, lean on mature enterprise search — bibryam · 2026-09-11
- Law Professor on Legal Engineering Jobs: Stigma Is Real but Builder Skills Open New Doors — jkubicki · 2026-09-11
- AI safety community mocked as 'bridge engineers' who say bridges can never be safe — Dan_Jeffries1 · 2026-09-11
- SoftBank's Masayoshi Son predicts 100 trillion self-replicating AIs: "humans' era as top life form is ending" — Puzzleheaded-King584 · 2026-09-11
- Inside Ant Group's play at WAIC-style expo: AI and hardware vendors settle into new division of labor — 智东西 · 2026-09-11
- Instagram head says engagement falls by half without the algorithm — hsuduebc2 · 2026-09-11