AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

An Optimal Control View of LoRA and Binary Controller Design for Vision Transformers

Chi Zhang, Jingpu Cheng, Qianxiao Li

ECCV 2024 · Architectures and efficiency › Pruning and efficient tuning · approach: Efficient ViT adaptation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace full-precision adaptation matrices with binary controls at rank 64AdaptFormer-64 · CIFAR-100accuracy: 85.90 -> 85.74No clear effect
replace full-precision adaptation matrices with binary controls at rank 64AdaptFormer-64 · SVHNaccuracy: 96.89 -> 97.01Improved
binary controls at rank 768 instead of full precisionAdaptFormer-768 · Food-101accuracy: 88.39 -> 88.31No clear effect
use high-rank binary control instead of rank-1 full-precision under same computeAdaptFormer-1 · SVHNaccuracy: 93.04 -> 97.03Improved
use high-rank binary control instead of rank-4 full-precision under same computeAdaptFormer-4 · Food-101accuracy: 85.42 -> 87.87Improved
reduce AdaptFormer rank to 1 (ablation)AdaptFormer · Food-101accuracy: drop of 6.45%Hurt
use high-rank binary control (BiC-3072) in place of full-tuning MLP layersViT MLP layers · CIFAR-100accuracy: marginally lower, difference of 0.11%No clear effect
use binary controllers in distributed trainingAdaptFormer-768 · SVHN (5 nodes)communication cost: 3.5% of full-precision costImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy