In product design, effective design space exploration (DSE) is crucial for generating high-quality design ideas, requiring designers to possess broad knowledge and balance various constraints. As large-scale models thrive, AI has become an indispensable design collaborator by providing cross-domain knowledge and assistance with complex reasoning. To facilitate collaborative DSE between designers and AI, we frame and advance the design process through the problem-solution co-evolution model and design reasoning methods. A formative study was conducted to identify key strategies for the implementation. Then we developed CoExploreDS, a system that formalizes problems and solutions emerging in the human-AI collaborative design space into nodes. Using four reasoning methods, this system dynamically generates suggestions based on the ongoing design process. User studies confirmed that CoExploreDS significantly improves design quality and the human-AI collaboration experience.
https://dl.acm.org/doi/10.1145/3706598.3713869
The ACM CHI Conference on Human Factors in Computing Systems (https://chi2025.acm.org/)