English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Knowledge graphs (KGs) consist of links that describe relationships betweenentities. Due to the difficulty of manually enumerating all relationshipsbetween entities, automatically completing them is essential for KGs. KnowledgeGraph Completion (KGC) is a task that infers unseen relationships betweenentities in a KG. Traditional embedding-based KGC methods, such as RESCAL,TransE, DistMult, ComplEx, RotatE, HAKE, HousE, etc., infer missing links usingonly the knowledge from training data. In contrast, the recent Pre-trainedLanguage Model (PLM)-based KGC utilizes knowledge obtained during pre-training.Therefore, PLM-based KGC can estimate missing links between entities by reusingmemorized knowledge from pre-training without inference. This approach isproblematic because building KGC models aims to infer unseen links betweenentities. However, conventional evaluations in KGC do not consider inferenceand memorization abilities separately. Thus, a PLM-based KGC method, whichachieves high performance in current KGC evaluations, may be ineffective inpractical applications. To address this issue, we analyze whether PLM-based KGCmethods make inferences or merely access memorized knowledge. For this purpose,we propose a method for constructing synthetic datasets specified in thisanalysis and conclude that PLMs acquire the inference abilities required forKGC through pre-training, even though the performance improvements mostly comefrom textual information of entities and relations.

Does Pre-trained Language Model Actually Infer Unseen Links in Knowledge Graph Completion?