ATWL: A Formal Language for Representing, Comparing, and Reusing Visual Analytics Workflows

The paper introduces ATWL, a declarative language that formalizes visual analytics workflows by specifying typed artifacts and transforms with defined analytical intents and actors. A library of 54 workflows, curated from published papers with LLM assistance, demonstrates patterns such as human-machine labor division and refinement cycles. Experiments show that using ATWL and the library improves LLM-designed workflows, though branching remains a limitation requiring syntax extension.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

• llms