Yan Zuo

Researcher, Pluralis Research

2 papers at NeurIPS 2025

OpenReview· Semantic Scholar· Google Scholar

Poster Session 4

Thursday, December 4, 2025 · 4:30 PM → 7:30 PM

Unextractable Protocol Models: Collaborative Training and Inference without Weight Materialization

#803 · Alexander Long, Chamin P Hewa Koneputugodage, Thalaiyasingam Ajanthan, Yan Zuo, Gil Avraham, Violetta Shevchenko, Hadi Mohaghegh Dolatabadi, Sameera Ramasinghe

We provide a decentralized framework for collaborators to run large model computation (training/inference) without any collaborator getting full access to the model.

Poster Session 6

1 paper

Friday, December 5, 2025 · 4:30 PM → 7:30 PM

Exhibit Hall C,D,E

Subspace Networks: Scaling Decentralized Training with Communication-Efficient Model Parallelism

#811 · Sameera Ramasinghe, Thalaiyasingam Ajanthan, Gil Avraham, Yan Zuo, Alexander Long

We propose a method to perform compression in pipeline parallel setting.