Page 179 - Read Online

Page 4 of 22 Hei et al. J. Mater. Inf. 2026, 6, 15

Figure 1. Workflow and framework of the proposed model for extracting and allocating entities. The upper section of the figure presents
the workflow, beginning with the retrieval of full-text research articles from Elsevier, followed by the construction of a specialized corpus.
Sentences are then extracted and annotated to obtain a JSON-formatted dataset, and the process concludes with model training and
inference. The model is primarily composed of two components: entity extraction and entity allocation. (A) Entity Extraction: This
component integrates MatSciBERT and a pointer network. MatSciBERT first tokenizes the input sentence and generates vector
representations for each token. The pointer network then computes the probability of each token serving as the head or tail of a specific
entity, thereby identifying entities based on these probabilities; (B) Entity Allocation: This component assesses whether entities of
different types belong to the same tuple using an entity matching score matrix. During model inference, the matching likelihood of entities
in corresponding order can be enhanced by multiplying the diagonal elements of the matrix by a parameter; (C) Entity Matching Score
Matrix: Each element of the matrix represents a combination of six vectors. The first two vectors correspond to the vector representations
of the two entities, while the remaining four vectors are derived from two attention mechanisms: intra-entity and inter-entity attention.
Intra-entity attention focuses on the attention distribution among different types of entities, whereas inter-entity attention concentrates on
the attention distribution within the same type of entity; (D) Inference: An example of four-tuple extraction is illustrated.

Preprocessing and annotation
In this stage, sentences describing significant alloy properties were identified from the corpus, and all tuples
within these sentences were annotated with scientifically accurate meanings as gold labels. Five entity types
were defined.