FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387164994> ?p ?o ?g. }

• W4387164994 abstract "Training text-to-image models with web scale image-text pairs enables the generation of a wide range of visual concepts from text. However, these pre-trained models often face challenges when it comes to generating highly aesthetic images. This creates the need for aesthetic alignment post pre-training. In this paper, we propose quality-tuning to effectively guide a pre-trained model to exclusively generate highly visually appealing images, while maintaining generality across visual concepts. Our key insight is that supervised fine-tuning with a set of surprisingly small but extremely visually appealing images can significantly improve the generation quality. We pre-train a latent diffusion model on $1.1$ billion image-text pairs and fine-tune it with only a few thousand carefully selected high-quality images. The resulting model, Emu, achieves a win rate of $82.9%$ compared with its pre-trained only counterpart. Compared to the state-of-the-art SDXLv1.0, Emu is preferred $68.4%$ and $71.3%$ of the time on visual appeal on the standard PartiPrompts and our Open User Input benchmark based on the real-world usage of text-to-image models. In addition, we show that quality-tuning is a generic approach that is also effective for other architectures, including pixel diffusion and masked generative transformer models." @default.
• W4387164994 created "2023-09-30" @default.
• W4387164994 creator A5001072947 @default.
• W4387164994 creator A5006043629 @default.
• W4387164994 creator A5009851374 @default.
• W4387164994 creator A5015536490 @default.
• W4387164994 creator A5019111856 @default.
• W4387164994 creator A5020152678 @default.
• W4387164994 creator A5027023070 @default.
• W4387164994 creator A5027576668 @default.
• W4387164994 creator A5035885005 @default.
• W4387164994 creator A5037049534 @default.
• W4387164994 creator A5037373363 @default.
• W4387164994 creator A5038297233 @default.
• W4387164994 creator A5038939197 @default.
• W4387164994 creator A5050342343 @default.
• W4387164994 creator A5054105523 @default.
• W4387164994 creator A5057482975 @default.
• W4387164994 creator A5057613852 @default.
• W4387164994 creator A5057748562 @default.
• W4387164994 creator A5064825062 @default.
• W4387164994 creator A5068664439 @default.
• W4387164994 creator A5071030446 @default.
• W4387164994 creator A5072331753 @default.
• W4387164994 creator A5076888060 @default.
• W4387164994 creator A5077105480 @default.
• W4387164994 creator A5080137972 @default.
• W4387164994 creator A5084862447 @default.
• W4387164994 date "2023-09-27" @default.
• W4387164994 modified "2023-09-30" @default.
• W4387164994 title "Emu: Enhancing Image Generation Models Using Photogenic Needles in a Haystack" @default.
• W4387164994 doi "https://doi.org/10.48550/arxiv.2309.15807" @default.
• W4387164994 hasPublicationYear "2023" @default.
• W4387164994 type Work @default.
• W4387164994 citedByCount "0" @default.
• W4387164994 crossrefType "posted-content" @default.
• W4387164994 hasAuthorship W4387164994A5001072947 @default.
• W4387164994 hasAuthorship W4387164994A5006043629 @default.
• W4387164994 hasAuthorship W4387164994A5009851374 @default.
• W4387164994 hasAuthorship W4387164994A5015536490 @default.
• W4387164994 hasAuthorship W4387164994A5019111856 @default.
• W4387164994 hasAuthorship W4387164994A5020152678 @default.
• W4387164994 hasAuthorship W4387164994A5027023070 @default.
• W4387164994 hasAuthorship W4387164994A5027576668 @default.
• W4387164994 hasAuthorship W4387164994A5035885005 @default.
• W4387164994 hasAuthorship W4387164994A5037049534 @default.
• W4387164994 hasAuthorship W4387164994A5037373363 @default.
• W4387164994 hasAuthorship W4387164994A5038297233 @default.
• W4387164994 hasAuthorship W4387164994A5038939197 @default.
• W4387164994 hasAuthorship W4387164994A5050342343 @default.
• W4387164994 hasAuthorship W4387164994A5054105523 @default.
• W4387164994 hasAuthorship W4387164994A5057482975 @default.
• W4387164994 hasAuthorship W4387164994A5057613852 @default.
• W4387164994 hasAuthorship W4387164994A5057748562 @default.
• W4387164994 hasAuthorship W4387164994A5064825062 @default.
• W4387164994 hasAuthorship W4387164994A5068664439 @default.
• W4387164994 hasAuthorship W4387164994A5071030446 @default.
• W4387164994 hasAuthorship W4387164994A5072331753 @default.
• W4387164994 hasAuthorship W4387164994A5076888060 @default.
• W4387164994 hasAuthorship W4387164994A5077105480 @default.
• W4387164994 hasAuthorship W4387164994A5080137972 @default.
• W4387164994 hasAuthorship W4387164994A5084862447 @default.
• W4387164994 hasBestOaLocation W43871649941 @default.
• W4387164994 hasConcept C115961682 @default.
• W4387164994 hasConcept C13280743 @default.
• W4387164994 hasConcept C13424479 @default.
• W4387164994 hasConcept C153180895 @default.
• W4387164994 hasConcept C154945302 @default.
• W4387164994 hasConcept C15744967 @default.
• W4387164994 hasConcept C167966045 @default.
• W4387164994 hasConcept C177264268 @default.
• W4387164994 hasConcept C185798385 @default.
• W4387164994 hasConcept C199360897 @default.
• W4387164994 hasConcept C205649164 @default.
• W4387164994 hasConcept C2780767217 @default.
• W4387164994 hasConcept C31972630 @default.
• W4387164994 hasConcept C36464697 @default.
• W4387164994 hasConcept C39890363 @default.
• W4387164994 hasConcept C41008148 @default.
• W4387164994 hasConcept C542102704 @default.
• W4387164994 hasConceptScore W4387164994C115961682 @default.
• W4387164994 hasConceptScore W4387164994C13280743 @default.
• W4387164994 hasConceptScore W4387164994C13424479 @default.
• W4387164994 hasConceptScore W4387164994C153180895 @default.
• W4387164994 hasConceptScore W4387164994C154945302 @default.
• W4387164994 hasConceptScore W4387164994C15744967 @default.
• W4387164994 hasConceptScore W4387164994C167966045 @default.
• W4387164994 hasConceptScore W4387164994C177264268 @default.
• W4387164994 hasConceptScore W4387164994C185798385 @default.
• W4387164994 hasConceptScore W4387164994C199360897 @default.
• W4387164994 hasConceptScore W4387164994C205649164 @default.
• W4387164994 hasConceptScore W4387164994C2780767217 @default.
• W4387164994 hasConceptScore W4387164994C31972630 @default.
• W4387164994 hasConceptScore W4387164994C36464697 @default.
• W4387164994 hasConceptScore W4387164994C39890363 @default.
• W4387164994 hasConceptScore W4387164994C41008148 @default.
• W4387164994 hasConceptScore W4387164994C542102704 @default.
• W4387164994 hasLocation W43871649941 @default.
• W4387164994 hasOpenAccess W4387164994 @default.
• W4387164994 hasPrimaryLocation W43871649941 @default.

• next