📄️ What is Human-LLM Interaction?
Given the growing capabilities of LLMs and the launch of ChatGPT in late 2022, human-LLM interaction has rapidly emerged as an important branch of human-AI interaction. Although much research has started to reference "human-LLM interaction," a precise definition of this term remains vague.
📄️ What is Human-LLM Interaction Mode?
As people increasingly use LLMs to accomplish a wide range of tasks, researchers in AI and HCI have developed tools to support these human-LLM interactions. For instance, designing customized prompts, uploading knowledge, and integrating these into OpenAI’s GPTs interface exemplifies the creation of a simple yet tailored LLM-powered system. In this case, while OpenAI’s model remains the core component—processing text inputs and generating responses—the system achieves enhanced functionality beyond its default capabilities. Similarly, various systems have been developed to expand human-LLM interactions, enabling more complex tasks than those typically associated with basic LLM usage.
📄️ What Large Language Models we are focusing on?
While broad generative AI can be used to build AI systems, our paper focuses on LLMs, which have garnered significant attention in recent years given their potential to augment the ability of humans to perform tasks across a wide variety of domains. As such, we consider the following three types of models: