AI Paper Atlas

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

Mitigating Semantic Collapse in Generative Personalization with Test-Time Embedding Adjustment

Anh Tuan Bui, Thuy-Trang Vu, Trung Le, Junae Kim, Tamas Abraham, Rollin Omari, Amardeep Kaur, Dinh Phung

ICLR 2026 · Image generation › Text-to-image diffusion · approach: Multi-subject personalization · 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
add test-time embedding adjustment (TEA)DreamBooth · CC101CLIP-T full prompt: +1.87Improved
add TEATextual Inversion · CC101CLIP-T full prompt: +0.64Improved
add TEATextual Inversion · CC101CLIP-I: -2.34Hurt
add TEADreamBooth · CelebACLIP-I: -2.37Hurt
add TEADreamBooth · CC101CLIP-I: +4.57Improved
add TEACustom Diffusion · CelebADINO-I: +3.09Improved
add TEACustom Diffusion / ClassDiffusion · CC101CLIP-T prompt-part: -0.13 / -0.12No clear effect
add TEAEasyControl · CC101 Pet DogDINO-I: 43.71 -> 48.32Improved
add TEAOminiControl · Subject PenguinVLM-P: 86.25 -> 90.50Improved
add TEAReVersion · Relationship Carved byCLIP-T full prompt: 27.74 -> 30.17Improved
increase rotation factor alpha (ablation)TEA · CC101 DreamBoothCLIP-T full prompt: 26.5 -> 28.2Improved
increase rotation factor alpha (ablation)TEA · CC101 DreamBoothCLIP-I: 73.5 -> 71.1Hurt

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