AI Paper Atlas

Research map / Architectures and efficiency / Mixture of experts

Training-Free Task Classification for Multi-Task Model Merging

Jungyong Son, Jinwook Jung, Sungyong Baik

ECCV 2026 · Architectures and efficiency › Mixture of experts · approach: Mixture-of-experts LoRA · 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
replace Euclidean distance with SVD projection residual for task classification (ablation)SiM · 8 vision tasks (CLIP ViT-B/32)task classification accuracy: 95.8 -> 99.0Improved
replace projection residual with Mahalanobis distance (ablation)SiM · 8 vision taskstask classification accuracy: 99.0 -> 96.5Hurt
replace projection residual with 1-NN (ablation)SiM · 8 vision taskstask classification accuracy: 99.0 -> 92.0Hurt
add SiM routing to compressed task vectors (TM-TA)TALL-Mask + Task Arithmetic · 20 vision tasks, ViT-B/32avg accuracy: 60.5 (Task Arithmetic) -> 89.6Improved
use SiM with larger task sets (30 and 50 tasks)TSV-C + SiM · 50 vision tasks, ViT-B/32avg accuracy: 80.2 vs best baseline 66.6 (TSV-M)Improved
SiM with ViT-L/14 on 8 tasks, TM-TA variantTM-TA + SiM · 8 vision tasks, ViT-L/14avg accuracy: 92.0 vs 94.9 (MoW-Merging) and 95.4 (TSV-C + SiM)Mixed
use SiM on unseen tasks (manifolds from 8 seen tasks only)SiM-merged models · unseen 6 vision tasksavg accuracy: seen 92.7 -> unseen 63.3 (TM-TIES); EMR 90.7 -> 57.9Mixed
add SiM routing on NLP tasksEMR/TM-TA/TM-TIES/TSV-C + SiM · 7 NLP tasks, T5-largeavg score: 88.8 fine-tuned vs 85.6 (TM-TIES + SiM); 79.6 TSV-MImproved
use SiM instead of per-sample dynamic interpolation (inference speed)SiM with group batch inference · 8 vision tasks, ViT-B/32latency per input: 0.63s (DaWin) -> 0.003sImproved
use 32 support samples per task instead of full training set (ablation)SiM · task classification, visiontask classification accuracyNo clear effect
replace SiM with oracle routing (ablation)SiM · vision downstreamavg accuracy: 0.3-0.6% gapNo clear effect

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