AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

Decoupled Multi-Predictor Optimization for Inference-Efficient Model Tuning

Liwei Luo, Shuaitengyuan Li, Dongwei Ren, Qilong Wang, Pengfei Zhu, Qinghua Hu

ICCV 2025 · Architectures and efficiency › Pruning and efficient tuning · approach: Adapter-based parameter-efficient tuning · 1 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
use DMPO (bypass modules, high-order predictors, two-phase loss weights) instead of Dyn-Adapter at 30% FLOPsearly-exit PEFT on ViT-B/16 · CIFAR-100 + 5 FGVC (average)average accuracy: 81.67 -> 88.08Improved
add multi-predictor architecture (bypass + high-order predictor) (ablation)DMPO · CIFAR-100 + 5 FGVC, 30% FLOPsaverage accuracy: 81.67 -> 85.16Improved
add decoupled optimization (two-phase loss weights) (ablation)DMPO · CIFAR-100 + 5 FGVC, 30% FLOPsaverage accuracy: 81.67 -> 84.41Improved
combine MPA and DO (ablation)DMPO · CIFAR-100 + 5 FGVC, 30% FLOPsaverage accuracy: 85.16 -> 88.08Improved
add MPA or DO at 70% FLOPs (ablation)DMPO · CIFAR-100 + 5 FGVC, 70% FLOPsaverage accuracy: 86.80 -> 86.98 (MPA), 86.96 (DO)No clear effect
use R2D order (small early weights first, large later) instead of D2R (ablation)decoupled optimization · CIFAR-100 + 5 FGVC, 30% FLOPsaverage accuracy: 86.43 -> 88.08Improved
use constant loss weights (early<deep or early>deep) instead of two-phase R2D (ablation)decoupled optimization · CIFAR-100 + 5 FGVC, 30% FLOPsaverage accuracy: 88.08 -> 85.03 and 83.86Hurt
add Sigmoid(loss) inter-stage control factor (ablation)DMPO · CIFAR-100 + 5 FGVC, 30% FLOPsaverage accuracy: 87.85 -> 88.08Improved
use Softmax or Tanh in place of Sigmoid for the control factor (ablation)DMPO · CIFAR-100 + 5 FGVC, 30% FLOPsaverage accuracy: 88.08 -> 80.69 (Tanh), 74.57 (Softmax)Hurt
DMPO at 30% FLOPs versus token pruning DyT at 70% FLOPs on VTAB-1KDMPO · VTAB-1Kaverage accuracy: 76.5 (DyT 0.7) vs 76.0 (DMPO 0.3)Mixed
DMPO instead of DyT on ImageNet-Sketch at 70% FLOPsDMPO · ImageNet-Sketch domain generalizationaccuracy: 25.7 -> 21.5Hurt

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