Zero-Shot Relation Extraction via Reading Comprehension
Explore this paper's citation graph
Summary
It is shown that relation extraction can be reduced to answering simple reading comprehension questions, by associating one or more natural-language questions with each relation slot, and that zero-shot generalization to unseen relation types is possible, at lower accuracy levels.
- Type
- preprint
- Published
- 2017-06-13
- Cited by
- 870
- References
- 22
- Access
- Open access
- OpenAlex
- https://openalex.org/W2624677889
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:793385
Keywords
Relation (database), Relationship extraction, Reading comprehension, Computer science, Generalization
References
- Large-scale Simple Question Answering with Memory Networks
- Teaching Machines to Read and Comprehend
- Wikidata: a new platform for collaborative data collection
- Open Information Extraction from the Web
- Knowledge-Based Weak Supervision for Information Extraction of Overlapping Relations
- GloVe: Global Vectors for Word Representation
- Representing Text for Joint Embedding of Text and Knowledge Bases
- Seed-Based Event Trigger Labeling: How far can event descriptions get us?
- Question-Answer Driven Semantic Role Labeling: Using Natural Language to Annotate Natural Language
- Injecting Logical Background Knowledge into Embeddings for Relation Extraction
- Lifted Rule Injection for Relation Embeddings
- Generalizing to Unseen Entities and Entity Pairs with Row-less Universal Schema
- Effective Crowd Annotation for Relation Extraction
- WikiReading: A Novel Large-scale Language Understanding Task over Wikipedia
- Learning Recurrent Span Representations for Extractive Question Answering
- Learning to Answer Questions from Wikipedia Infoboxes
- Multi-Perspective Context Matching for Machine Comprehension
- Bidirectional Attention Flow for Machine Comprehension
- SQuAD: 100,000+ Questions for Machine Comprehension of Text
- End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures
Cited by
- Zero-Shot Question Generation from Knowledge Graphs for Unseen Predicates and Entity Types
- Neural architectures for open-type relation argument extraction
- code2vec: learning distributed representations of code
- Know What You Don’t Know: Unanswerable Questions for SQuAD
- Question Answering Resources Applied to Slot-Filling
- Zero-Shot Relation Extraction from Word Embeddings
- Factoid Question Answering with Distant Supervision
- Zero-shot visual recognition via latent embedding learning
- The Natural Language Decathlon: Multitask Learning as Question Answering
- QuAC: Question Answering in Context
- Interpretation of Natural Language Rules in Conversational Machine Reading
- Crowdsourcing Semantic Label Propagation in Relation Classification
- pair2vec: Compositional Word-Pair Embeddings for Cross-Sentence Inference
- U-Net: Machine Reading Comprehension with Unanswerable Questions
- Building Dynamic Knowledge Graphs from Text using Machine Reading Comprehension
- Playing by the Book: Towards Agent-based Narrative Understanding through Role-playing and Simulation
- Reduce, Reuse, Recycle: New uses for old QA resources
- Hybrid Attention-Based Prototypical Networks for Noisy Few-Shot Relation Classification
- Multi-style Generative Reading Comprehension
- A Survey of Zero-Shot Learning
Related papers
- SQuAD: 100,000+ Questions for Machine Comprehension of Text
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
- TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension
- Deep Contextualized Word Representations
- Adam: A Method for Stochastic Optimization
- Attention is All you Need
- GloVe: Global Vectors for Word Representation
- Long Short-Term Memory