AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

LoRA-S: An Efficient Low Rank Adaptation scheme via Sylvester equation

Jinyang ZHENG, Tong Wu

ICLR 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: LoRA variants and theory · 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
lift Adam to horizontal space (AdamS, efficient feature learning)Adam · Mix-of-Show (r=4)CLIP score: 25.17 -> 32.20Improved
lift Adam to horizontal space (AdamS, efficient feature learning)Adam · Mix-of-Show (r=4)FID: 72.75 -> 54.39Improved
use AdamS instead of Scaled GDScaled GD · Mix-of-Show (r=16)CLIP score: 29.07 -> 32.64Improved
use AdamS instead of Scaled GDScaled GD · Mix-of-Show (r=16)FID: 59.87 -> 46.32Improved
use AdamS instead of LoRA-RiteLoRA-Rite · Mix-of-Show (r=4)CLIP score: 30.96 -> 32.20Improved
use AdamS instead of AdamAdam · E2E NLG GPT-2 medium (r=4)BLEU: 68.0 -> 69.4Improved
use LRACS instead of AdamWAdamW · E2E NLG GPT-2 medium (r=4)BLEU: 68.6 -> 70.4Improved
use LRACS instead of LoRA-RiteLoRA-Rite · E2E NLG GPT-2 medium (r=4)BLEU: 69.3 -> 70.4Improved
use AdamS instead of AdamAdam · E2E NLG GPT-2 small (r=4)BLEU: 68.8 -> 69.4Improved
add weight-decay regularization of the Sylvester matrix K (ablation)AdamS · Mix-of-Show (r=16)CLIP score: 32.64 -> 31.97Hurt
increase rank from 4 to 16 (ablation)AdamS · Mix-of-ShowFID: 54.39 -> 46.32Improved
vary learning rate from 50% to 250% of default (ablation)AdamS · Mix-of-Show (r=4)CLIP score: 32.11 -> 31.95No clear effect

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