Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

VeRA: Vector-based Random Matrix Adaptation, Dawid J. Kopiczko+, N/A, arXiv'23 #1211

Open
AkihikoWatanabe opened this issue Jan 17, 2024 · 0 comments

Comments

@AkihikoWatanabe
Copy link
Owner

AkihikoWatanabe commented Jan 17, 2024

URL

Affiliations

  • Dawid J. Kopiczko, N/A
  • Tijmen Blankevoort, N/A
  • Yuki M. Asano, N/A

Abstract

  • Low-rank adapation (LoRA) is a popular method that reduces the number oftrainable parameters when finetuning large language models, but still facesacute storage challenges when scaling to even larger models or deployingnumerous per-user or per-task adapted models. In this work, we presentVector-based Random Matrix Adaptation (VeRA), which significantly reduces thenumber of trainable parameters compared to LoRA, yet maintains the sameperformance. It achieves this by using a single pair of low-rank matricesshared across all layers and learning small scaling vectors instead. Wedemonstrate its effectiveness on the GLUE and E2E benchmarks, imageclassification tasks, and show its application in instruction-tuning of 7B and13B language models.

Translation (by gpt-3.5-turbo)

  • 低ランク適応(LoRA)は、大規模な言語モデルのfine-tuning時に訓練可能なパラメータの数を減らす人気のある手法ですが、さらに大きなモデルや多数のユーザーやタスクごとの適応モデルを展開する際には、ストレージの課題が深刻です。
    本研究では、ベクトルベースのランダム行列適応(VeRA)を提案し、LoRAと比較して訓練可能なパラメータの数を大幅に削減しながら、同じ性能を維持します。
    これは、すべてのレイヤーで共有される1つのペアの低ランク行列を使用し、代わりに小さなスケーリングベクトルを学習することによって実現されます。
    GLUEとE2Eのベンチマーク、画像分類タスクでの効果を示し、7Bおよび13Bの言語モデルのインストラクションチューニングへの応用も示します。

Summary (by gpt-3.5-turbo)

  • 本研究では、大規模な言語モデルのfine-tuningにおいて、訓練可能なパラメータの数を削減するための新しい手法であるベクトルベースのランダム行列適応(VeRA)を提案する。
    VeRAは、共有される低ランク行列と小さなスケーリングベクトルを使用することで、同じ性能を維持しながらパラメータ数を削減する。
    GLUEやE2Eのベンチマーク、画像分類タスクでの効果を示し、言語モデルのインストラクションチューニングにも応用できることを示す。
@AkihikoWatanabe AkihikoWatanabe changed the title VeRA: Vector-based Random Matrix Adaptation, Dawid J. Kopiczko+, N/A, arXiv'23 Jan 17, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant