# From Research Question to Scientific Workflow: Leveraging Agentic AI for Science Automation

arXiv: https://arxiv.org/abs/2604.21910 · HTML: https://mlsanity.com/paper/2604.21910/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 3 of its 16 references that are in the map.
- First posted on arXiv 23 Apr 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 3 of 3)

- [Towards Scientific Intelligence: A Survey of LLM-based Scientific Agents](https://mlsanity.com/paper/2503.24047/) (2025)
- [A Terminology for Scientific Workflow Systems](https://mlsanity.com/paper/2506.07838/) (2025)
- [A Community Roadmap for Scientific Workflows Research and Development](https://mlsanity.com/paper/2110.02168/) (2021)

## Built on by (most cited first, 3 of 3)

- [DataFlow-Harness: A Grounded Code-Agent Platform for Constructing Editable LLM Data Pipelines](https://mlsanity.com/paper/2607.16617/) (2026)
- [Complexity and Scale in AI-Assisted Workflow Management: A Federated Learning Case Study](https://mlsanity.com/paper/2606.18425/) (2026)
- [An Agentic Interface for End-to-End Probabilistic Seismic Hazard and Risk Analysis](https://mlsanity.com/paper/2607.16249/) (2026)

## Abstract (arXiv)

Scientific workflow systems automate execution -- scheduling, fault tolerance, resource management -- but not the semantic translation that precedes it. Scientists still manually convert research questions into workflow specifications, a task requiring both domain knowledge and infrastructure expertise. We propose an agentic architecture that closes this gap through three layers: an LLM interprets natural language into structured intents (semantic layer); validated generators produce reproducible workflow DAGs (deterministic layer); and domain experts author ``Skills'': markdown documents encoding vocabulary mappings, parameter constraints, and optimization strategies (knowledge layer). This decomposition confines LLM non-determinism to intent extraction: identical intents always yield identical workflows. We implement and evaluate the architecture on the 1000 Genomes population genetics workflow and Hyperflow WMS running on Kubernetes. In an ablation study on 150 queries, Skills raise full-match intent accuracy from 44% to 83%; skill-driven deferred workflow generation reduces data transfer by 92%; and the end-to-end pipeline completes queries on Kubernetes with LLM overhead below 15 seconds and cost under 0.001 per query.