[3rd CFP] Call for contributions: Structuring scientific process parameters

26 views
Skip to first unread message

Jennifer D'Souza

unread,
Jan 24, 2026, 2:49:00 AM (yesterday) Jan 24
to Machine Learning News

Dear colleagues,

We invite contributions to a community-driven effort to structure how experimental and simulation processes are described in research papers, across applied natural sciences and engineering.

Many ML applications rely on scientific datasets, yet the underlying processes (e.g., PCR, ALD/CVD, tensile testing, soil leaching, fMRI tasks) are typically described only in free text and in highly inconsistent ways. This makes comparison, reproducibility, FAIR reuse, and ML-based reasoning over methods extremely difficult.

Our goal is straightforward:
from collections of papers describing the same process, we extract and refine a shared, structured description of process parameters and conditions (e.g., inputs, settings, tools, outputs), together with domain experts. These descriptions will be published openly as ORKG templates and form the basis of a planned Nature Scientific Data paper.

How you can contribute

  • Provide a collection of full-text papers (~50+) describing a specific experimental or simulation process, and/or

  • Give expert feedback on the extracted parameter structure

Individual or small-team participation is welcome; co-authorship opportunities are available.

Example processes (non-exhaustive):
https://docs.google.com/document/d/1iyL1l9vCXhnQ0To7j79vlr-pW4JvPlQC95svygqRDfg/edit?usp=sharing

👉 Register your interest: https://forms.gle/9WEdouw4yMyNHcn19

This is the third and final call for participation in this effort.
Selected contributors will be notified by Jan 31, 2026; data collection runs until Apr 30, 2026.

Best regards,
Jennifer D’Souza
TIB Hannover
(on behalf of the schema-miner coordination team)


Reply all
Reply to author
Forward
0 new messages