Kiroong Choe

Astrid van den Brandt, Kiroong Choe, Sehi L'Yi et al.

Diverse genomics data, scientific questions, and analysis tasks typically demand highly specialized visualizations. Therefore, users often must customize or author new ones tailored to their data. Existing tools are usually either limited in customization or require substantial learning or programming, and even expressive tools assume visualization expertise many users lack. Agentic and large language model (LLM) approaches are increasingly applied to complex scientific tasks, including visualization. Natural-language conversational interfaces offer a promising path to democratizing the authoring of complex visualizations. In the context of genomics, these approaches face additional challenges: genomics visualizations typically integrate heterogeneous data types and are composed of multiple linked interactive views. These challenges motivate more structured LLM-based schemes. We first characterize where vanilla LLM generation succeeds and fails for genomics visualization, identifying eight quality dimensions. We then compare six schemes--direct generation, a fixed pipeline, and four agentic configurations varying in the number of specialist agents and the presence of a reviewer--across 159 cases spanning three levels of query ambiguity and specification complexity. All schemes use the Gosling visualization grammar as structured output. Agentic iteration substantially improves perceived quality over both baselines, while more complex agent architectures yield no additional benefit. We discuss implications for designing agentic systems for domain-specific visualization authoring. All supplemental materials are available at https://osf.io/uqe83.

Jiyeon Bae, Mingyu An, Jeongin Park et al.

Exploratory data analysis (EDA) is often hindered by cold-start friction; when users lack specific analytic goals, they struggle to configure complex visualization parameters. While existing visualization tools mostly rely on explicit user input to frame data, we propose leveraging the physical environment as an implicit framing mechanism. We introduce a conceptual framework that uses the geometric and spatial properties of physical containers in Augmented Reality (AR) to guide data interpretation. We characterize how container attributes, such as number of faces, size, proportion, and shape, give rise to distinct perceptual tendencies. For example, a circular container may encourage cyclic interpretation, while juxtaposed planar faces may facilitate comparative analysis. By treating physical forms as environmental framing conditions, we show how AR can orient a user's attention and structure their exploration without requiring manual encoding or prescribing fixed conclusions. We demonstrate this framework through a series of AR design examples illustrating how container morphology foregrounds cyclic, comparative, and sequential analytic patterns.

Youngtaek Kim, Hyeon Jeon, Kiroong Choe et al.

Software developers explore and inspect software repository data to obtain detailed information archived in the development history. However, developers who are not acquainted with the development context suffer from delving into the repositories with a handful of information; they have difficulty discovering and expanding information fragments considering the topological and sequential multi-dimensional structure of repositories. We introduce ExIF, an interactive visualization for exploring information fragments in software repositories. ExIF helps users discover new information fragments within clusters or topological neighbors and identify revisions incorporating user-collected fragments.