{"ID":2879885,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.15610","arxiv_id":"2508.15610","title":"Transduction is All You Need for Structured Data Workflows","abstract":"This paper introduces Agentics, a functional agentic AI framework for building LLM-based structured data workflow pipelines. Designed for both research and practical applications, Agentics offers a new data-centric paradigm in which agents are embedded within data types, enabling logical transduction between structured states. This design shifts the focus toward principled data modeling, providing a declarative language where data types are directly exposed to large language models and the data values are composed through transductions between input and output types. We present a range of structured data workflow tasks and empirical evidence demonstrating the effectiveness of this approach, including data wrangling, text-to-SQL semantic parsing, and domain-specific multiple-choice question answering, and data-driven scientific discovery tasks.","short_abstract":"This paper introduces Agentics, a functional agentic AI framework for building LLM-based structured data workflow pipelines. Designed for both research and practical applications, Agentics offers a new data-centric paradigm in which agents are embedded within data types, enabling logical transduction between structured...","url_abs":"https://arxiv.org/abs/2508.15610","url_pdf":"https://arxiv.org/pdf/2508.15610v3","authors":"[\"Alfio Gliozzo\",\"Naweed Khan\",\"Christodoulos Constantinides\",\"Nandana Mihindukulasooriya\",\"Nahuel Defosse\",\"Gaetano Rossiello\",\"Junkyu Lee\"]","published":"2025-08-21T14:35:47Z","proceeding":"cs.AI","tasks":"[\"cs.AI\",\"cs.LG\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}