Explainability-in-Action: Enabling Expressive Manipulation and Tacit Understanding by Bending Diffusion Models in ComfyUI
This addresses the need for artistic engagement and modifiability in creative AI applications, though it appears incremental as it builds on existing interfaces and model-bending concepts.
The paper tackles the problem of making large-scale generative models like text-to-image diffusion systems more accessible and controllable for artists by proposing a craft-based explainability approach that enables interactive manipulation of model components through a plugin in ComfyUI, demonstrating that artists can develop intuition about how each part influences outputs.
Explainable AI (XAI) in creative contexts can go beyond transparency to support artistic engagement, modifiability, and sustained practice. While curated datasets and training human-scale models can offer artists greater agency and control, large-scale generative models like text-to-image diffusion systems often obscure these possibilities. We suggest that even large models can be treated as creative materials if their internal structure is exposed and manipulable. We propose a craft-based approach to explainability rooted in long-term, hands-on engagement akin to Schön's "reflection-in-action" and demonstrate its application through a model-bending and inspection plugin integrated into the node-based interface of ComfyUI. We demonstrate that by interactively manipulating different parts of a generative model, artists can develop an intuition about how each component influences the output.