Compositional Inversion for Stable Diffusion Models
Authors: Xulu Zhang, Xiao-Yong Wei, Jinlin Wu, Tianyi Zhang, Zhaoxiang Zhang, Zhen Lei, Qing Li
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experimental results demonstrate the effectiveness of our proposed approach in mitigating the overfitting problem and generating more diverse and balanced compositions of concepts in the synthesized images. |
| Researcher Affiliation | Academia | Xulu Zhang1,2, Xiao-Yong Wei3,1*, Jinlin Wu2,5, Tianyi Zhang1, Zhaoxiang Zhang2,4,5, Zhen Lei2,4,5, Qing Li1 1Department of Computing, the Hong Kong Polytechnic University, Hong Kong 2Center for Artificial Intelligence and Robotics, HKISI, CAS, Hong Kong 3College of Computer Science, Sichuan University, Chengdu, China 4School of Artificial Intelligence, UCAS, Beijing, China 5State Key Laboratory of Multimodal Artificial Intelligence Systems, CASIA, Beijing, China |
| Pseudocode | No | The paper does not contain any explicit pseudocode or algorithm blocks. |
| Open Source Code | Yes | The source code is available at https://github.com/zhangxulu1996/Compositional-Inversion. |
| Open Datasets | Yes | To test the generalizability, we generate prompts by combining the inverted concepts with 80 categories from the COCO dataset (Lin et al. 2014) using the conjunction word and . |
| Dataset Splits | No | The paper does not explicitly provide training/test/validation dataset splits or specific percentages/counts for each split. |
| Hardware Specification | No | The paper does not explicitly describe the hardware specifications used to run its experiments. |
| Software Dependencies | No | The paper mentions software like Stable Diffusion and Hugging Face, but does not provide specific version numbers for these or other key software dependencies. |
| Experiment Setup | No | The paper does not provide specific details about hyperparameters, training configurations, or other system-level settings for the experimental setup. |