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 (57)

• Recent Advances in Natural Language Processing via Large Pre-Trained Language Models: A Survey
• Structured Prediction as Translation between Augmented Natural Languages
• A Frustratingly Easy Approach for Entity and Relation Extraction
• Learning from Noisy Labels for Entity-Centric Information Extraction
• Fine-grained Video-Text Retrieval with Hierarchical Graph Reasoning
• TexSmart: A Text Understanding System for Fine-Grained NER and Enhanced Semantic Analysis
• Reasoning Over Semantic-Level Graph for Fact Checking
• DARE: Data Augmented Relation Extraction with GPT-2
• PharmKE: Knowledge Extraction Platform for Pharmaceutical Texts using Transfer Learning
• "When they say weed causes depression, but it's your fav antidepressant": Knowledge-aware Attention Framework for Relationship Extraction
• Temporal Common Sense Acquisition with Minimal Supervision
• Knowledge Enhanced Contextual Word Representations
• Efficient long-distance relation extraction with DG-SpanBERT
• Unsupervised Label-aware Event Trigger and Argument Classification
• Language Models as Emotional Classifiers for Textual Conversations
• Document-Level Event Argument Extraction by Conditional Generation
• Semantic Role Labeling as Dependency Parsing: Exploring Latent Tree Structures Inside Arguments
• "Is depression related to cannabis?": A knowledge-infused model for Entity and Relation Extraction with Limited Supervision
• Czert -- Czech BERT-like Model for Language Representation
• HANet: Hierarchical Alignment Networks for Video-Text Retrieval
• Hierarchical Deep Residual Reasoning for Temporal Moment Localization
• EXTRACTOR: Extracting Attack Behavior from Threat Reports
• A Heterogeneous Graph with Factual, Temporal and Logical Knowledge for Question Answering Over Dynamic Contexts
• WebRED: Effective Pretraining And Finetuning For Relation Extraction On The Web
• Dialogue-Based Relation Extraction
• Extracting a Knowledge Base of Mechanisms from COVID-19 Papers
• Improving Portuguese Semantic Role Labeling with Transformers and Transfer Learning
• Improving Robustness by Augmenting Training Sentences with Predicate-Argument Structures
• Human-like Controllable Image Captioning with Verb-specific Semantic Roles
• An MRC Framework for Semantic Role Labeling
• Towards Theme Detection in Personal Finance Questions
• A Benchmark for Structured Procedural Knowledge Extraction from Cooking Videos
• RDSGAN: Rank-based Distant Supervision Relation Extraction with Generative Adversarial Framework
• Speaker attribution in German parliamentary debates with QLoRA-adapted large language models
• 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
• Semantic Role Labeling Guided Multi-turn Dialogue ReWriter
• Interactive Machine Comprehension with Dynamic Knowledge Graphs
• NLPBK at VLSP-2020 shared task: Compose transformer pretrained models for Reliable Intelligence Identification on Social network
• Inconsistent Few-Shot Relation Classification via Cross-Attentional Prototype Networks with Contrastive Learning
• Robust Question Answering Through Sub-part Alignment
• CSAGN: Conversational Structure Aware Graph Network for Conversational Semantic Role Labeling
• Within-Document Event Coreference with BERT-Based Contextualized Representations
• Asking It All: Generating Contextualized Questions for any Semantic Role
• Commonsense Knowledge in Word Associations and ConceptNet
• Hierarchical Multitask Learning with Dependency Parsing for Japanese Semantic Role Labeling Improves Performance of Argument Identification
• Few Shot Learning for Information Verification
• Meta Sequence Learning for Generating Adequate Question-Answer Pairs
• Grounding Implicit Goal Description for Robot Indoor Navigation Via Recursive Belief Update
• Building a Hebrew Semantic Role Labeling Lexical Resource from Parallel Movie Subtitles
• Semi-Automating Knowledge Base Construction for Cancer Genetics
• Conversational Semantic Role Labeling
• Fine-Grained Chemical Entity Typing with Multimodal Knowledge Representation
• SRLGRN: Semantic Role Labeling Graph Reasoning Network
• Framing Unpacked: A Semi-Supervised Interpretable Multi-View Model of Media Frames