AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

DeCoFlow: Structural Decomposition of Normalizing Flows for Continual Anomaly Detection

hun im, Jungi Lee, Subeen Cha, Pilsung Kang

ECCV 2026 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection · 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
decompose coupling subnets into frozen base + LoRA instead of other architectures with frozen+LoRA (ablation)normalizing flow · MVTec-ADI-AUC: 62.03-68.15 -> 98.47Improved
decompose at feature level instead of coupling level (ablation)DeCoFlow NF · MVTec-ADI-AUC: 98.47 -> 50.00Hurt
remove ACL (ablation)DeCoFlow · MVTec-ADI-AUC: -10.17 %p when removedImproved
add Tail-Aware Loss (ablation)DeCoFlow · MVTec-ADI-AUC: +3.17 %p (full model vs without TAL)Improved
add Task-Specific Alignment (ablation)DeCoFlow · MVTec-ADI-AUC: +0.68 %p (full vs without TSA)Improved
use LoRA instead of full per-task subnet (ablation)DeCoFlow · MVTec-ADI-AUC: -0.04 %p when LoRA excluded; doubles per-task paramsNo clear effect
apply TAL when base is frozen vs trainable (ablation)normalizing flow · MVTec-ADI-AUC: Frozen +13.04 %p; Trainable -1.22 %pMixed
add TSA under trainable base (ablation)normalizing flow · MVTec-ADI-AUC: -16.34 %p (trainable); -1.19 %p (frozen)Hurt
replace ACL with additional DCL blocks (8DCL+0ACL) (ablation)DeCoFlow · MVTec-ADI-AUC/performance: -9.01 %pHurt
increase LoRA rank from 16 to 128 (ablation)DeCoFlow · MVTec-ADI-AUC / P-AP: 0.02 / 0.09 %p changeNo clear effect
update shared base or use high-rank adapters (ablation)DeCoFlow · MVTec-ADI-AUC: 4.7-25.0x parameter cost, no significant gainsNo clear effect
force 5% misrouting of tasks (ablation)DeCoFlow · MVTec-ADP-AP: 56.93 -> 30.20Hurt

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