Infrastructure · Computer science · General science

data-to-paper links LLM research automation to traceable manuscripts

Traceable LLM-agent workflow from data analysis to manuscript drafting.

Summary

The arXiv paper presents data-to-paper, an automation platform that guides interacting LLM agents through hypothesis generation, research planning, code writing and debugging, result interpretation and manuscript drafting while tracing information flow for human verification.

AI role

Guided agents through hypothesis generation, planning, coding, result interpretation, and paper drafting while tracing information flow.

Narrative role

data-to-paper supports the evidence ledger by making verifiability part of automated research workflows.

Caveat

Traceability helps auditing, but it does not prove that generated analyses are novel, correct, or important.