3 papers
cs.LG2026
Aurora: A Leverage-Aware Spectral Optimizer
Alec Dewulf, Dhruv Pai, Li Yang +2
We show that for tall matrix parameters, like projection matrices in the MLP layers, the Muon update can have row norms that are arbitrarily non-uniform. This can lead to a self-re…
cs.LG2026
Parallax: Parameterized Local Linear Attention for Language Modeling
Yifei Zuo, Dhruv Pai, Zhichen Zeng +3
Large Language Models (LLMs) have become the central paradigm in artificial intelligence, yet the core computational primitive of attention has remained structurally unchanged. Loc…
cs.LG2026
PreFT: Prefill-only finetuning for efficient inference
Andrew Lanpouthakoun, Aryaman Arora, Zhengxuan Wu +4
Large language models can now be personalised efficiently at scale using parameter efficient finetuning methods (PEFTs), but serving user-specific PEFTs harms throughput, even with…