AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

From Parameter to Representation: A Closed-Form Approach for Controllable Model Merging

Jialin Wu, Jian Yang, Handing Wang, Jiajun Wen, Zhiyong Yu

AAAI 2026 · Learning theory and algorithms › Combinatorial and search optimization · approach: Multi-objective optimization · 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 ReACT correction on AMPP instead of plain AMPPAdaMerging++ · 8 tasks ViT-B/32average accuracy: 85.4 vs. 81.1Improved
use ReACT instead of Pareto Merging under one-hot preferenceAMPP · 8 tasks ViT-B/32 one-hotaverage accuracy: 88.9% vs. 83.6%Improved
use ReACT instead of Pareto Merging under equal preferenceAM · 8 tasks ViT-B/32 equal prefaverage accuracy: 84.9 vs 84.9No clear effect
use ReACT instead of Pareto Merging on controllabilityAM · Cars/RESISC45/DTD priorityHV@8: 60.90 -> 68.49Improved
use ReACT instead of Pareto Merging on controllabilityAMPP · Cars/RESISC45/DTD priorityUniformity U@8: 26.59 -> 41.58Improved
restrict calibration data to 10% of unlabeled test dataReACT (AMPP) · 8 tasks one-hotaverage accuracy: 88.9 -> 86.7Hurt
add orthogonal regularization (beta>0 vs beta=0) (ablation)ReACT · 8 tasks ViT-B/32accuracyImproved
use regularized solution instead of pure orthogonal transformation (beta to infinity) (ablation)ReACT · 8 tasks ViT-B/32accuracyImproved
vary beta across a wide range (ablation)ReACT · 8 tasks ViT-B/32accuracyNo clear effect
use data-aware Pareto-optimal aggregation instead of naive averaging (ablation)ReACT · equal preferenceNAcc: 92.2 -> 93.5Improved
use data-aware Pareto-optimal aggregation instead of polar decomposition (ablation)ReACT · equal preferenceNAcc: 91.5 -> 93.5Improved
use data-aware aggregation instead of naive averaging (ablation)ReACT · priority preferenceHV: 67.6 -> 70.2Improved

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