LoRMA: Low-Rank Multiplicative Adaptation for LLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bihany, Harsh, Patel, Shubham, Modi, Ashutosh
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912419477979136
author Bihany, Harsh
Patel, Shubham
Modi, Ashutosh
author_facet Bihany, Harsh
Patel, Shubham
Modi, Ashutosh
contents Large Language Models have shown remarkable capabilities in the NLP domain. Their effectiveness can mainly be attributed to their ability to adapt to an array of downstream tasks. However, generally, full fine-tuning is a computationally expensive job. To mitigate this, many techniques have been developed that prime efficiency, a prominent one being Low-Rank Adaptation (LoRA). However, LoRA and its variants employ re-parametrized additive updates. In this paper, we propose Low-Rank Multiplicative Adaptation (LoRMA), which shifts the paradigm of additive updates to a richer space of matrix multiplicative transformations. We tackle challenges such as computational complexity and rank bottleneck of matrix multiplication by effectively re-ordering operations and introducing rank inflation strategies. We conduct extensive experiments to demonstrate the effectiveness of our approach in terms of various evaluation metrics.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07621
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LoRMA: Low-Rank Multiplicative Adaptation for LLMs
Bihany, Harsh
Patel, Shubham
Modi, Ashutosh
Computation and Language
Artificial Intelligence
Machine Learning
Large Language Models have shown remarkable capabilities in the NLP domain. Their effectiveness can mainly be attributed to their ability to adapt to an array of downstream tasks. However, generally, full fine-tuning is a computationally expensive job. To mitigate this, many techniques have been developed that prime efficiency, a prominent one being Low-Rank Adaptation (LoRA). However, LoRA and its variants employ re-parametrized additive updates. In this paper, we propose Low-Rank Multiplicative Adaptation (LoRMA), which shifts the paradigm of additive updates to a richer space of matrix multiplicative transformations. We tackle challenges such as computational complexity and rank bottleneck of matrix multiplication by effectively re-ordering operations and introducing rank inflation strategies. We conduct extensive experiments to demonstrate the effectiveness of our approach in terms of various evaluation metrics.
title LoRMA: Low-Rank Multiplicative Adaptation for LLMs
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2506.07621