AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

InvCoSS: Inversion-driven Continual Self-supervised Learning in Medical Multi-modal Image Pre-training

Zihao Luo, Shaohao Rui, Zhenyu Tang, Guotai Wang, Xiaosong Wang

CVPR 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised neuroimaging representations · 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
add total-variation image prior Limg (ablation)InvCoSS · six downstream tasksAVG up: 86.12 -> 86.75Improved
add total-variation image prior Limg (ablation)InvCoSS · six downstream tasksAVG down: 27.12 -> 24.29Improved
add repulsive representation loss Lrep (ablation)InvCoSS · six downstream tasksAVG up: 86.75 -> 88.03Improved
add repulsive representation loss Lrep (ablation)InvCoSS · RICORDACC: 82.14 -> 84.52Improved
add Lrep (ablation)InvCoSS · six downstream tasksAVG down: 24.29 -> 20.11Improved
use InvUNet instead of bottom-to-top generator (ablation)InvCoSS · ChestXR (generator comparison)ACC: 90.77 -> 92.27Improved
use InvUNet instead of larger bottom-to-top generator (ablation)InvCoSS · ChestXR (generator comparison)ACC: 91.21 -> 92.27Improved
add generator G (InvUNet) vs direct image optimization (ablation)InvCoSS · QaTaDSC: 74.42 -> 76.23Improved
add generator G for 3D data (ablation)InvCoSS · CT and MRI tasksfeasibilityImproved
replace data replay with inversion-synthesized imagesInvCoSS vs ER · nine downstream tasksAVG up: 85.45 -> 89.17Improved
replace data replay with inversion-synthesized imagesInvCoSS vs MedCoSS · nine downstream tasksAVG up: 89.03 -> 89.17No clear effect
use InvCoSS instead of data-free CaSSLeCSSL · nine downstream tasksAVG up: 86.88 -> 89.17Improved

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