Simple BERT Models for Relation Extraction and Semantic Role Labeling

Abstract

We present simple BERT-based models for relation extraction and semantic role labeling. In recent years, state-of-the-art performance has been achieved using neural models by incorporating lexical and syntactic features such as part-of-speech tags and dependency trees. In this paper, extensive experiments on datasets for these two tasks show that without using any external features, a simple BERT-based model can achieve state-of-the-art performance. To our knowledge, we are the first to successfully apply BERT in this manner. Our models provide strong baselines for future research.

work in progress

References in corpus (2)

• Simplifying Graph Convolutional Networks
• Improving Relation Extraction by Pre-trained Language Representations

Cited by in corpus (38)

• Recent Advances in Natural Language Processing via Large Pre-Trained Language Models: A Survey
• Fine-grained Video-Text Retrieval with Hierarchical Graph Reasoning
• TexSmart: A Text Understanding System for Fine-Grained NER and Enhanced Semantic Analysis
• DARE: Data Augmented Relation Extraction with GPT-2
• Knowledge Enhanced Contextual Word Representations
• Temporal Common Sense Acquisition with Minimal Supervision
• Document-Level Event Argument Extraction by Conditional Generation
• Language Models as Emotional Classifiers for Textual Conversations
• "Is depression related to cannabis?": A knowledge-infused model for Entity and Relation Extraction with Limited Supervision
• HANet: Hierarchical Alignment Networks for Video-Text Retrieval
• WebRED: Effective Pretraining And Finetuning For Relation Extraction On The Web
• A Heterogeneous Graph with Factual, Temporal and Logical Knowledge for Question Answering Over Dynamic Contexts
• EXTRACTOR: Extracting Attack Behavior from Threat Reports
• Hierarchical Deep Residual Reasoning for Temporal Moment Localization
• Dialogue-Based Relation Extraction
• Human-like Controllable Image Captioning with Verb-specific Semantic Roles
• RDSGAN: Rank-based Distant Supervision Relation Extraction with Generative Adversarial Framework
• Towards Theme Detection in Personal Finance Questions
• Experiments on transfer learning architectures for biomedical relation extraction
• Semantic Graphs for Generating Deep Questions
• Controllable Open-ended Question Generation with A New Question Type Ontology
• Topic-Aware Evidence Reasoning and Stance-Aware Aggregation for Fact Verification
• NLPBK at VLSP-2020 shared task: Compose transformer pretrained models for Reliable Intelligence Identification on Social network
• Semantic Role Labeling Guided Multi-turn Dialogue ReWriter
• Inconsistent Few-Shot Relation Classification via Cross-Attentional Prototype Networks with Contrastive Learning
• Interactive Machine Comprehension with Dynamic Knowledge Graphs
• CSAGN: Conversational Structure Aware Graph Network for Conversational Semantic Role Labeling
• Fine-Grained Chemical Entity Typing with Multimodal Knowledge Representation
• Framing Unpacked: A Semi-Supervised Interpretable Multi-View Model of Media Frames
• Few Shot Learning for Information Verification
• Hierarchical Multitask Learning with Dependency Parsing for Japanese Semantic Role Labeling Improves Performance of Argument Identification
• Building a Hebrew Semantic Role Labeling Lexical Resource from Parallel Movie Subtitles
• Semi-Automating Knowledge Base Construction for Cancer Genetics
• SRLGRN: Semantic Role Labeling Graph Reasoning Network
• Grounding Implicit Goal Description for Robot Indoor Navigation Via Recursive Belief Update
• Conversational Semantic Role Labeling
• Commonsense Knowledge in Word Associations and ConceptNet
• Asking It All: Generating Contextualized Questions for any Semantic Role