Texterial: A Text-as-Material Interaction Paradigm for LLM-Mediated Writing

2026-02-27

AI Summary

Current generative AI writing tools feature linear and rigid interfaces that fail to harness the rich textual manipulation capabilities enabled by large language models (LLMs), thereby constraining users’ creative expression. This work proposes a “text-as-material” interaction paradigm, systematically introducing material metaphors—such as clay and plants—into LLM-based writing interactions for the first time. These metaphors encourage users to co-create through sculpting or gardening-like processes, fundamentally reshaping both writing workflows and mental models. Building on metaphor-driven interaction design, gesture-based shaping, and time-evolving generation mechanisms, the authors developed two technical probes: “text-as-clay” and “text-as-plant.” Formative studies demonstrate that users intuitively grasp and effectively leverage these metaphors, significantly enhancing creative immersion and exploratory behavior, thereby expanding the design space for AI-assisted writing tools.

Technology Category

• Humans and AI: Game Design — Procedural Content Generation & Storytelling
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: Generation

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

What if text could be sculpted and refined like clay -- or cultivated and pruned like a plant? Texterial reimagines text as a material that users can grow, sculpt, and transform. Current generative-AI models enable rich text operations, yet rigid, linear interfaces often mask such capabilities. We explore how the text-as-material metaphor can reveal AI-enabled operations, reshape the writing process, and foster compelling user experiences. A formative study shows that users readily reason with text-as-material, informing a conceptual framework that explains how material metaphors shift mental models and bridge gulfs of envisioning, execution, and evaluation in LLM-mediated writing. We present the design and evaluation of two technical probes: Text as Clay, where users refine text through gestural sculpting, and Text as Plants, where ideas grow serendipitously over time. This work expands the design space of writing tools by treating text as a living, malleable medium.

Problem

• text-as-material
• LLM-mediated writing
• user interface
• mental models
• generative AI

Innovation

• text-as-material
• LLM-mediated writing
• material metaphor
• generative AI interaction
• writing interface design