Back
Reparameterized LLM Training via Orthogonal Equivalence Transformation
@conference{Qiuetal25b, title = {Reparameterized LLM Training via Orthogonal Equivalence Transformation}, booktitle = {Advances in Neural Information Processing Systems 38 (NeurIPS 2025)}, month = dec, year = {2025}, author = {Qiu, Z. and Buchholz, S. and Xiao, T. and Dax, M. and Sch{\"o}lkopf, B. and Liu, W.}, month_numeric = {12} }
More information