AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

GLOBA: Rethinking Parameter Conflicts in Model Merging

Zehao Liu, Kun Li, Wei Zhou

AAAI 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 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 GLOBA instead of Task Arithmetic when merging Math and Code7B merge · MBPP (Math+Code, 7B)pass@1 (%): 15.00 -> 18.80Improved
use GLOBA when merging Instruct and Code7B merge · MBPP (Instruct+Code, 7B)pass@1 (%): 20.80 -> 31.60Improved
use GLOBA instead of Task Arithmetic when merging Math and Code13B merge · MBPP (Math+Code, 13B)pass@1 (%): 8.80 -> 21.80Improved
use GLOBA instead of TIES-Merging when merging Math and Code13B merge · MBPP (Math+Code, 13B)pass@1 (%): 15.60 -> 21.80Improved
use GLOBA instead of TA when merging Instruct and Code13B merge · AlpacaEval (Instruct+Code, 13B)win rate (%): 77.97 -> 82.12Improved
use GLOBA instead of TIES when merging Math and Code (GSM8K)7B merge · GSM8K (Math+Code, 7B)accuracy (%): 4.32 -> 51.48Improved
use GLOBA instead of TIES for merging Instruct and Math7B merge · AlpacaEval (Instruct+Math, 7B)win rate (%): 39.93 -> 39.14No clear effect
add completely orthogonal (Type A) patterns after pruning (ablation)pruned base model · Math+Instruct merge, 7BGSM8K accuracy (%): 49.20 -> 50.27No clear effect
add same-sign overlap (Type D+) patterns to WizardMath-7B (ablation)WizardMath-7B · GSM8K (Math+Code, 7B)accuracy (%): 51.63 -> 5.38Hurt
add opposite-sign (Type D-) patterns to WizardMath-7B (ablation)WizardMath-7B · GSM8K (Math+Code, 7B)accuracy (%): 51.63 -> 0Hurt
add all overlap patterns for Math+Code merge (ablation)WizardMath-7B · GSM8K (Math+Code, 7B)accuracy (%): 51.63 -> 43.14Hurt
add all patterns for Instruct into Math model (ablation)WizardMath-7B · AlpacaEval (Math+Instruct, 7B)win rate (%): 13.00 -> 38.57Improved

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