The field of human-computer interaction is witnessing significant developments with the integration of large language models (LLMs). Recent research has focused on leveraging LLMs to enhance personalized interactions, improve task-oriented decision-making, and facilitate natural language understanding. A notable direction is the use of LLMs as design materials for interactive applications, such as museum installations and robot trainers, to create more engaging and dynamic user experiences. Additionally, researchers are exploring the potential of LLMs in addressing challenges like object hallucinations in vision-language models and developing cognitive architectures for co-constructive task learning. Noteworthy papers include Agent-Centric Personalized Multiple Clustering with Multi-Modal LLMs, which proposes a framework for personalized clustering using MLLMs, and DASH: Detection and Assessment of Systematic Hallucinations of VLMs, which introduces a pipeline for identifying systematic hallucinations in vision-language models. These advancements demonstrate the potential of LLMs to revolutionize human-computer interaction and enable more effective, personalized, and adaptive interfaces.