GPart is a PEFT technique that achieves end-to-end isometry by using a single random isometric partition matrix to map a low-dimensional trainable vector into full model weights without low-rank structure.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
GPart: End-to-End Isometric Fine-Tuning via Global Parameter Partitioning
GPart is a PEFT technique that achieves end-to-end isometry by using a single random isometric partition matrix to map a low-dimensional trainable vector into full model weights without low-rank structure.