Ziye Ma

Over-parameterization's impact on robustness is not monotonic; solution geometry and optimization path, not just parameter count, dictate sensitivity in matrix sensing.

The optimal LoRA rank is not intrinsic to the model but dynamically dictated by the data's geometric properties, enabling principled data selection for efficient fine-tuning.

MALT outperforms Muon in pretraining language models while keeping memory usage and computation time nearly identical.