AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

DA-MergeLoRA: Hypernetwork-Based LoRA Merging for Few-Shot Test-Time Domain Adaptation

Siobhan Reid, Zhixiang Chi, Li Gu, Omid Heidari, Ziqiang Wang, Yang Wang

ECCV 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Conditional distribution alignment · 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
merge per-domain LoRAs with a meta-learned hypernetwork instead of using prior FSTT-DA method L2CCLIP ViT-B/16 · DomainNet averageaccuracy: 61.2 -> 62.44Improved
use hypernetwork merging instead of L2C on FMoW worst-caseCLIP ViT-B/16 · FMoW OODWC accuracy: 40.9 -> 52.26Improved
use hypernetwork merging instead of L2C on iWildCam F1CLIP ViT-B/16 · iWildCam OODmacro F1: 35.2 -> 33.62Hurt
use hypernetwork merging instead of per-domain best on QuickdrawCLIP ViT-B/16 · DomainNet Quickdrawaccuracy: 18.7 -> 17.87Hurt
use learned hypernetwork merge weights instead of simple parameter averaging (ablation)source LoRAs · FMoW OODWC accuracy: 20.63 -> 52.26Improved
use hypernetwork instead of few-shot AdaMerging of LoRAs (ablation)source LoRAs · iWildCam OODmacro F1: 18.66 -> 33.62Improved
use hypernetwork merging instead of simple averaging on a saturated dataset (ablation)source LoRAs · Camelyon17 OODaccuracy: 95.26 -> 96.90Improved
use per-domain LoRAs with merging instead of one universal LoRA (ablation)CLIP ViT-B/16 · FMoW OODWC accuracy: 51.02 -> 52.26Improved
condition hypernetwork on target images instead of noise (ablation)hypernetwork · iWildCam OODmacro F1: 17.39 -> 33.62Improved
add domain tokens to LoRA columns (ablation)hypernetwork · iWildCam OODmacro F1: 25.72 -> 33.62Improved
add layer and layer-type tokens (ablation)hypernetwork · FMoW OODWC accuracy: 49.16 -> 52.26Improved
remove architectural components (tokens, target images) on Camelyon17 (ablation)hypernetwork · Camelyon17 OODaccuracy: 96.92 vs 96.90No 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