collaborators

6 papers

cs.LG2026

Dion3: Full-Stack Orthogonal Updates

Noah Amsel, Jack Zhang, Kwangjun Ahn +5

The Muon optimizer incurs a significant overhead cost due to its cubic-time Newton-Schulz orthogonalization step. When weights are sharded, communication overhead compounds this co…

cs.LG2026

Next-Latent Prediction Transformers Learn Compact World Models

Jayden Teoh, Manan Tomar, Kwangjun Ahn +7

Transformers replace recurrence with a memory that grows with sequence length and self-attention that enables ad-hoc lookups over past tokens. Consequently, they lack an inherent i…

cs.LG2025

Dion2: A Simple Method to Shrink Matrix in Muon

Kwangjun Ahn, Noah Amsel, John Langford

The Muon optimizer enjoys strong empirical performance and theoretical grounding. However, the super-linear cost of its orthonormalization step introduces increasing overhead with…

cs.LG2025

The Belief State Transformer

Edward S. Hu, Kwangjun Ahn, Qinghua Liu +9

We introduce the "Belief State Transformer", a next-token predictor that takes both a prefix and suffix as inputs, with a novel objective of predicting both the next token for the…

cs.LG2025

Dion: Distributed Orthonormalized Updates

Kwangjun Ahn, Byron Xu, Natalie Abreu +5

Orthonormalized updates accelerate training, improve stability, and enable robust hyperparameter transfer, but existing methods like Muon rely on dense matrix operations that clash…

cs.LG2025

Efficient Joint Prediction of Multiple Future Tokens

Kwangjun Ahn, Alex Lamb, John Langford

In this short report, we introduce joint multi-token prediction (JTP), a lightweight modification of standard next-token prediction designed to enrich hidden state representations…