Customized Text-to-Image Generation With Curriculum Multi-Reward Reinforcement Learning
- 1 January 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28, 13259-13270
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (62222209)
- National Key Research and Development Program of China (2022ZD0115903)
This publication has 31 references indexed in Scilit:
- Powerful and Flexible: Personalized Text-to-Image Generation via Reinforcement Learning2024
- InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning2024
- DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation2023
- Multi-Concept Customization of Text-to-Image Diffusion2023
- High-Resolution Image Synthesis with Latent Diffusion Models2022
- Curriculum Learning: A SurveyInternational Journal of Computer Vision, 2022
- Emerging Properties in Self-Supervised Vision Transformers2021
- CurriculumNet: Weakly Supervised Learning from Large-Scale Web Images2018
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Simple statistical gradient-following algorithms for connectionist reinforcement learningMachine Learning, 1992