TensorX
返回文献探索

Paper · arXiv 2310.11454

VeRA: Vector-based Random Matrix Adaptation

Dawid Jan Kopiczko, Tijmen Blankevoort, Yuki Markus Asano

31 upvotesOctober 17, 2023arXiv 预印本
AI 摘要

Vector-based Random Matrix Adaptation (VeRA) reduces the number of trainable parameters by 10x compared to LoRA while maintaining performance, and is demonstrated on benchmarks like GLUE and E2E, showing its utility in instruction-following.

Low-rank adaptationLoRAVector-based Random Matrix AdaptationVeRAGLUEE2Einstruction-followingLlama2

Abstract

Low-rank adapation (LoRA) is a popular method that reduces the number of trainable parameters when finetuning large language models, but still faces acute storage challenges when scaling to even larger models or deploying numerous per-user or per-task adapted models. In this work, we present Vector-based Random Matrix Adaptation (VeRA), which reduces the number of trainable parameters by 10x compared to LoRA, yet maintains the same performance. It achieves this by using a single pair of low-rank matrices shared across all layers and learning small scaling vectors instead. We demonstrate its effectiveness on the GLUE and E2E benchmarks, and show its application in instruction-following with just 1.4M parameters using the Llama2 7B model.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
VeRA: Vector-based Random Matrix Adaptation | TensorX