AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Beyond Pixel Mimicry: Disentangled Self-Similarity Rewards for Diverse Subject-Driven Generation

Qian Wang, Zhenyu Li, Abdelrahman Eldesokey, Peter Wonka

ECCV 2026 · Image generation › Text-to-image diffusion · approach: Multi-subject personalization

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 DivRL (nSSM + gated VSM) post-trainingFlux-Kontext · DreamBench++MTG-nSSM: 0.659 -> 0.689Improved
add DivRL post-trainingFlux-Kontext · DreamBench++DINO-nSSM: 0.433 -> 0.453Improved
add DivRL post-trainingFlux-Kontext · DreamBench++VSM-0.7: 0.605 -> 0.614No clear effect
add DivRL post-trainingFlux-Kontext · DreamBench++HPS: 0.294 -> 0.287Hurt
add nSSM diversity reward to VSM reward (ablation)DivRL (VSM only) · DreamBench++MTG-nSSM: 0.608 -> 0.689Improved
add nSSM diversity reward to VSM reward (ablation)DivRL (VSM only) · DreamBench++VSM-0.7: 0.688 -> 0.614Hurt
add VSM identity gate to nSSM-only reward (ablation)DivRL (nSSM only) · DreamBench++VSM-0.7: 0.562 -> 0.614Improved
add VSM identity gate to nSSM-only reward (ablation)DivRL (nSSM only) · DreamBench++MTG-nSSM: 0.704 -> 0.689Hurt
add VSM identity gate to nSSM-only reward (ablation)DivRL (nSSM only) · DreamBench++CLIP-T: 0.287 -> 0.279Hurt
use gated two-stage reward instead of linear weighting of VSM and nSSM (ablation)DivRL · DreamBench++consistency ratio and diversity-over-consistencyImproved
add 2x2 average pooling (48x48 -> 24x24) to features for nSSM (ablation)DivRL · DreamBench++visual artifactsImproved

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