arXiv Machine Learning By Jed A. Duersch, Na\"im Es-Sebbani, Nathana\"el Haas, Zied Bouraoui

Dense Structural Compression of Transformers via Gauge-Correct Channel Removal

Read the original on arXiv Machine Learning →

The paper introduces GaugeLasso, a method that applies symmetric group‑lasso penalties to transformer channels during training, enabling entire tensor slices to be zeroed out while maintaining dense tensors for GPU efficiency. By calibrating channel penalties based on inference utility per compute, the network self‑organizes into depth‑dependent structural profiles that can be dramatically smaller than the original architecture, achieving up to 255‑fold compression on a polynomial division task and outperforming hand‑designed baselines on language modeling and autoencoding benchmarks. The approach also accelerates training and reveals over‑provisioned axes that guide subsequent design iterations.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Computation and Language
Aug 31

Pruning Laws for Large Language Models

arXiv:2504.04342v2 Announce Type: replace Abstract: Scaling up model parameters and training data consistently improves the performance of large language models (LLMs), but at the cost of rapidly gro...

By Ayan Sengupta, Siddhant Chaudhary, Tanmoy Chakraborty