More Like this

1. Generating conceptual landscape design via text-to-image generative AI model

   https://doi.org/10.1177/23998083251316064  

   Ye, Xinyue; Huang, Tianchen; Song, Yang; Li, Xin; Newman, Galen; Wu, Dayong_Jason; Zeng, Yijun (January 2025, Environment and Planning B: Urban Analytics and City Science)

   This study explores the integration of text-to-image generative AI, particularly Stable Diffusion, in conjunction with ControlNet and LoRA models in conceptual landscape design. Traditional methods in landscape design are often time-consuming and limited by the designer’s individual creativity, also often lacking efficiency in the exploration of diverse design solutions. By leveraging AI tools, we demonstrate a workflow that efficiently generates detailed and visually coherent landscape designs, including natural parks, city plazas, and courtyard gardens. Through both qualitative and quantitative evaluations, our results indicate that fine-tuned models produce superior designs compared to non-fine-tuned models, maintaining spatial consistency, control over scale, and relevant landscape elements. This research advances the efficiency of conceptual design processes and underscores the potential of AI in enhancing creativity and innovation in landscape architecture. 

   more » « less
2. Towards Open Domain Text-Driven Synthesis of Multi-person Motions

   Shan, Mengyi; Dong, Lu; Han, Yutao; Yao, Yuan; Liu, Tao; Nwogu, Ifeoma; Qi, Guo_Jun; Hill, Mitch (October 2024, Springer_Science+Business_Media)

   This work aims to generate natural and diverse group motions of multiple humans from textual descriptions. While singleperson text-to-motion generation is extensively studied, it remains challenging to synthesize motions for more than one or two subjects from in-the-wild prompts, mainly due to the lack of available datasets. In this work, we curate human pose and motion datasets by estimating pose information from large-scale image and video datasets. Our models use a transformer-based diffusion framework that accommodates multiple datasets with any number of subjects or frames. Experiments explore both generation of multi-person static poses and generation of multiperson motion sequences. To our knowledge, our method is the first to generate multi-subject motion sequences with high diversity and fidelity from a large variety of textual prompts. 

   more » « less
3. How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?

   Bansal, Hritik; Yin, Da; Monajatipoor, Masoud; Chang, Kai-Wei (January 2022, Empirical Methods in Natural Language Processing)

   Text-to-image generative models have achieved unprecedented success in generating high-quality images based on natural language descriptions. However, it is shown that these models tend to favor specific social groups when prompted with neutral text descriptions (e.g., ‘a photo of a lawyer’). Following Zhao et al. (2021), we study the effect on the diversity of the generated images when adding ethical intervention that supports equitable judgment (e.g., ‘if all individuals can be a lawyer irrespective of their gender’) in the input prompts. To this end, we introduce an Ethical NaTural Language Interventions in Text-to-Image GENeration (ENTIGEN) benchmark dataset to evaluate the change in image generations conditional on ethical interventions across three social axes – gender, skin color, and culture. Through CLIP-based and human evaluation on minDALL.E, DALL.E-mini and Stable Diffusion, we find that the model generations cover diverse social groups while preserving the image quality. In some cases, the generations would be anti-stereotypical (e.g., models tend to create images with individuals that are perceived as man when fed with prompts about makeup) in the presence of ethical intervention. Preliminary studies indicate that a large change in the model predictions is triggered by certain phrases such as ‘irrespective of gender’ in the context of gender bias in the ethical interventions. We release code and annotated data at https://github.com/Hritikbansal/entigen_emnlp. 

   more » « less
4. DSPO: Direct Score Preference Optimization for Diffusion Model Alignment.

   Zhu, Huaisheng; Xiao, Teng; Honavar, Vasant G (July 2025, International Conference on Learning Representations (ICLR 2025))

   Diffusion-based Text-to-Image (T2I) models have achieved impressive success in generating high-quality images from textual prompts. While large language models (LLMs) effectively leverage Direct Preference Optimization (DPO) for fine-tuning on human preference data without the need for reward models, diffusion models have not been extensively explored in this area. Current preference learning methods applied to T2I diffusion models immediately adapt existing techniques from LLMs. However, this direct adaptation introduces an estimated loss specific to T2I diffusion models. This estimation can potentially lead to suboptimal performance through our empirical results. In this work, we propose Direct Score Preference Optimization (DSPO), a novel algorithm that aligns the pretraining and fine-tuning objectives of diffusion models by leveraging score matching, the same objective used during pretraining. It introduces a new perspective on preference learning for diffusion models. Specifically, DSPO distills the score function of human-preferred image distributions into pretrained diffusion models, fine-tuning the model to generate outputs that align with human preferences. We theoretically show that DSPO shares the same optimization direction as reinforcement learning algorithms in diffusion models under certain conditions. Our experimental results demonstrate that DSPO outperforms preference learning baselines for T2I diffusion models in human preference evaluation tasks and enhances both visual appeal and prompt alignment of generated images. 

   more » « less
5. Your Diffusion Model is Secretly a Zero-Shot Classifier

   Alexander C. Li, Mihir Prabhudesai (September 2023, ICCV)

   The recent wave of large-scale text-to-image diffusion models has dramatically increased our text-based image generation abilities. These models can generate realistic images for a staggering variety of prompts and exhibit impressive compositional generalization abilities. Almost all use cases thus far have solely focused on sampling; however, diffusion models can also provide conditional density estimates, which are useful for tasks beyond image generation. In this paper, we show that the density estimates from large-scale text-to-image diffusion models like Stable Diffusion can be leveraged to perform zero-shot classification without any additional training. Our generative approach to classification, which we call Diffusion Classifier, attains strong results on a variety of benchmarks and outperforms alternative methods of extracting knowledge from diffusion models. Although a gap remains between generative and discriminative approaches on zero-shot recognition tasks, our diffusion-based approach has significantly stronger multimodal compositional reasoning ability than competing discriminative approaches. Finally, we use Diffusion Classifier to extract standard classifiers from class-conditional diffusion models trained on ImageNet. Our models achieve strong classification performance using only weak augmentations and exhibit qualitatively better "effective robustness" to distribution shift. Overall, our results are a step toward using generative over discriminative models for downstream tasks. 

   more » « less