Generating Fine-Grained Open Vocabulary Entity Type Descriptions
This work addresses a specific need in knowledge graph completion by providing open-vocabulary descriptions for entities, though it appears incremental as it builds on existing methods for text generation.
The paper tackles the problem of generating short textual descriptions for entities in knowledge graphs that lack them, by introducing a dynamic memory-based network that leverages fact embeddings and generation context, and demonstrates improved accuracy against strong baselines.
While large-scale knowledge graphs provide vast amounts of structured facts about entities, a short textual description can often be useful to succinctly characterize an entity and its type. Unfortunately, many knowledge graph entities lack such textual descriptions. In this paper, we introduce a dynamic memory-based network that generates a short open vocabulary description of an entity by jointly leveraging induced fact embeddings as well as the dynamic context of the generated sequence of words. We demonstrate the ability of our architecture to discern relevant information for more accurate generation of type description by pitting the system against several strong baselines.