Publishing FAIR and Machine-actionable Reviews in Materials Science: The Case for Symbolic Knowledge in Neuro-symbolic Artificial Intelligence

TL;DR

The paper addresses the challenge of reusing ALD/ALE review insights that are trapped in PDFs by publishing machine-actionable, FAIR comparisons within the ORKG. It demonstrates converting nine ALD/ALE reviews into 18 ORKG Comparisons, building a dataset of 33 natural-language questions with SPARQL gold standards, and evaluating three QA regimes (SPARQL, PDF-based LLMs, and LLMs grounded in ORKG tables). Domain experts rate the machine-actionable tables as scientifically meaningful and practically useful, while neural QA over PDFs underperforms relative to symbolic querying; grounding LLMs in ORKG tables dramatically improves performance, though it does not fully replace symbolic querying. The results support a neurosymbolic paradigm where a symbolic core provides exact, reproducible answers, and neural models offer exploratory and explanatory capabilities when anchored to structured, curated data. The work advocates for broader community adoption of machine-actionable reviews and outlines a roadmap toward tighter symbolic–neural integration for robust scientific AI in materials science.

Abstract

Scientific reviews are central to knowledge integration in materials science, yet their key insights remain locked in narrative text and static PDF tables, limiting reuse by humans and machines alike. This article presents a case study in atomic layer deposition and etching (ALD/E) where we publish review tables as FAIR, machine-actionable comparisons in the Open Research Knowledge Graph (ORKG), turning them into structured, queryable knowledge. Building on this, we contrast symbolic querying over ORKG with large language model-based querying, and argue that a curated symbolic layer should remain the backbone of reliable neurosymbolic AI in materials science, with LLMs serving as complementary, symbolically grounded interfaces rather than standalone sources of truth.

Figures (11)

• Figure 1: Modeling a survey table as machine-actionable data in the Open Research Knowledge Graph (ORKG). Panel (1) shows Table 5 from the rare-earth ALD review by Ghazy et al. ghazy2025atomic. Panel (2) shows the corresponding ORKG comparison https://orkg.org/comparisons/R1469991, where rows of the original table are represented as contribution columns and the original table column headers as ORKG properties. For readability, only three of the nine papers included in the full comparison are displayed. The following seven are modeled properties as web resources: https://orkg.org/properties/P180031, https://orkg.org/properties/P180032, https://orkg.org/properties/P180033, https://orkg.org/properties/P180034, https://orkg.org/properties/P180035, https://orkg.org/properties/P180036, and https://orkg.org/properties/P180037.
• Figure 2: Continuation of \ref{['fig:table-to-comparison']}. Starts with Panel (1) which is the knowledge graph view (the actual graph representation of the machine-actionable Table 5 ghazy2025atomic). Panel 2 zooms in on the first contribution, linking the comparison entry to its individual paper record in the ORKG https://orkg.org/papers/R1469778/R1469970 (Panel 3), which corresponds to the first data row in Table 5 in the review by Ghazy et al ghazy2025atomic and the article "Polycrystalline Er-doped Y3Ga5O12 nanofilms fabricated by atomic layer deposition on silicon at a low temperature and the exploration on electroluminescence performance"yu2022polycrystalline.
• Figure 3: Methodology for importing survey tables into the Open Research Knowledge Graph (ORKG). Figure reproduced from Oelen et al. (2020), "Creating a Scholarly Knowledge Graph from Survey Article Tables" oelen2020kgfromsurveytable, where this technique was first introduced.
• Figure 4: Illustration of key features and anatomy of SmartReviews. They are composed of several building blocks, including natural text, comparisons, and visualizations. Figure reproduced from Oelen et al. (2021), "SmartReviews: Towards Human- and Machine-Actionable Representation of Review Articles" oelensmartreviews, where the SmartReview methodology was first introduced.
• Figure 5: Screenshot of the visual SPARQL editor provided on the ORKG platform (https://orkg.org/sparql/), showing the execution of the Q.1 for Table 2 from the research article "Saturation profile based conformality analysis for atomic layer deposition: aluminum oxide in lateral high-aspect-ratio channels"yim2020saturation, modeled in the ORKG as the machine-actionable Comparison https://orkg.org/comparisons/R1469158. The query retrieves co-occurrences of reactor types and LHAR structures across studies, illustrating how different experimental setups pair deposition systems with test geometries. The persistent query URL can be reproduced at https://tinyurl.com/lhr-reactor.