ORES: Open-vocabulary Responsible Visual Synthesis
This addresses the need for responsible AI in visual synthesis by enabling models to avoid diverse forbidden concepts based on user input, though it appears incremental as it builds on existing LLM and diffusion methods.
The paper tackles the problem of avoiding forbidden visual concepts in image synthesis by formalizing Open-vocabulary Responsible Visual Synthesis (ORES) and proposing a Two-stage Intervention (TIN) framework, which reduces risks in image generation as demonstrated experimentally.
Avoiding synthesizing specific visual concepts is an essential challenge in responsible visual synthesis. However, the visual concept that needs to be avoided for responsible visual synthesis tends to be diverse, depending on the region, context, and usage scenarios. In this work, we formalize a new task, Open-vocabulary Responsible Visual Synthesis (ORES), where the synthesis model is able to avoid forbidden visual concepts while allowing users to input any desired content. To address this problem, we present a Two-stage Intervention (TIN) framework. By introducing 1) rewriting with learnable instruction through a large-scale language model (LLM) and 2) synthesizing with prompt intervention on a diffusion synthesis model, it can effectively synthesize images avoiding any concepts but following the user's query as much as possible. To evaluate on ORES, we provide a publicly available dataset, baseline models, and benchmark. Experimental results demonstrate the effectiveness of our method in reducing risks of image generation. Our work highlights the potential of LLMs in responsible visual synthesis. Our code and dataset is public available.