Task-specific Pre-training And Prompt Decomposition For Knowledge Graph Population With Language Models

Li Tianyi, Huang Wenyu, Papasarantopoulos Nikos, Vougiouklis Pavlos, Pan Jeff Z.. Arxiv 2022

[Paper]    
Applications BERT Model Architecture Prompting Training Techniques

We present a system for knowledge graph population with Language Models, evaluated on the Knowledge Base Construction from Pre-trained Language Models (LM-KBC) challenge at ISWC 2022. Our system involves task-specific pre-training to improve LM representation of the masked object tokens, prompt decomposition for progressive generation of candidate objects, among other methods for higher-quality retrieval. Our system is the winner of track 1 of the LM-KBC challenge, based on BERT LM; it achieves 55.0% F-1 score on the hidden test set of the challenge.

Similar Work