Generative Interfaces for Language Models
This addresses the problem of inefficient human-AI interaction for users in multi-turn and exploratory tasks, representing a novel paradigm rather than an incremental improvement.
The paper tackles the inefficiency of linear request-response formats in LLM interactions by proposing Generative Interfaces for Language Models, which proactively generate UIs for adaptive engagement, resulting in up to a 72% improvement in human preference over traditional chat-based interfaces.
Large language models (LLMs) are increasingly seen as assistants, copilots, and consultants, capable of supporting a wide range of tasks through natural conversation. However, most systems remain constrained by a linear request-response format that often makes interactions inefficient in multi-turn, information-dense, and exploratory tasks. To address these limitations, we propose Generative Interfaces for Language Models, a paradigm in which LLMs respond to user queries by proactively generating user interfaces (UIs) that enable more adaptive and interactive engagement. Our framework leverages structured interface-specific representations and iterative refinements to translate user queries into task-specific UIs. For systematic evaluation, we introduce a multidimensional assessment framework that compares generative interfaces with traditional chat-based ones across diverse tasks, interaction patterns, and query types, capturing functional, interactive, and emotional aspects of user experience. Results show that generative interfaces consistently outperform conversational ones, with up to a 72% improvement in human preference. These findings clarify when and why users favor generative interfaces, paving the way for future advancements in human-AI interaction.