Zero-Shot Relation Extraction via Reading Comprehension

We show that relation extraction can be reduced to answering simple reading comprehension questions, by associating one or more natural-language questions with each relation slot. This reduction has several advantages: we can (1) learn relation-extraction models by extending recent neural reading-comprehension techniques, (2) build very large training sets for those models by combining relation-specific crowd-sourced questions with distant supervision, and even (3) do zero-shot learning by extracting new relation types that are only specified at test-time, for which we have no labeled training examples. Experiments on a Wikipedia slot-filling task demonstrate that the approach can generalize to new questions for known relation types with high accuracy, and that zero-shot generalization to unseen relation types is possible, at lower accuracy levels, setting the bar for future work on this task.

• Publication year

  2017

• Venue

  Conference on Computational Natural Language Learning

• Publication date

  2017-06-13

• Fields of study

  Computer Science

• Identifiers
  DOI 10.18653/v1/K17-1034arXiv 1706.04115
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Multi-Perspective Context Matching for Machine Comprehension

  2016cited by this paper
• Bidirectional Attention Flow for Machine Comprehension

  2016influential reference
• Dynamic Coattention Networks For Question Answering

  2016cited by this paper
• Effective Crowd Annotation for Relation Extraction

  2016cited by this paper
• SQuAD: 100,000+ Questions for Machine Comprehension of Text

  2016influential reference
• Generalizing to Unseen Entities and Entity Pairs with Row-less Universal Schema

  2016influential reference
• Learning to Answer Questions from Wikipedia Infoboxes

  2016influential reference
• End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures

  2016cited by this paper
• Lifted Rule Injection for Relation Embeddings

  2016cited by this paper
• WikiReading: A Novel Large-scale Language Understanding Task over Wikipedia

  2016influential reference
• Learning Recurrent Span Representations for Extractive Question Answering

  2016cited by this paper
• Question-Answer Driven Semantic Role Labeling: Using Natural Language to Annotate Natural Language

  2015cited by this paper
• Representing Text for Joint Embedding of Text and Knowledge Bases

  2015cited by this paper
• Seed-Based Event Trigger Labeling: How far can event descriptions get us?

  2015cited by this paper
• Teaching Machines to Read and Comprehend

  2015cited by this paper
• Injecting Logical Background Knowledge into Embeddings for Relation Extraction

  2015cited by this paper
• Large-scale Simple Question Answering with Memory Networks

  2015cited by this paper
• GloVe: Global Vectors for Word Representation

  2014cited by this paper
• Relation Extraction with Matrix Factorization and Universal Schemas

  2013cited by this paper
• Wikidata: a new platform for collaborative data collection

  2012cited by this paper
• Knowledge-Based Weak Supervision for Information Extraction of Overlapping Relations

  2011cited by this paper
• Open Information Extraction from the Web

  2007cited by this paper

• CrispEdit: Low-Curvature Projections for Scalable Non-Destructive LLM Editing

  2026influential citation
• Persuasion Tokens for Editing Factual Knowledge in LLMs

  2026cites this paper
• DCCL: Question-guided dual-channel contrastive learning framework for emotion-cause pair extraction

  2026cites this paper
• Reverse-Engineering Model Editing on Language Models

  2026influential citation
• Spectral Characterization and Mitigation of Sequential Knowledge Editing Collapse

  2026cites this paper
• A Study on Building Efficient Zero-Shot Relation Extraction Models

  2026cites this paper
• Conflict-Resolving and Sharpness-Aware Minimization for Generalized Knowledge Editing with Multiple Updates

  2026influential citation
• Beyond Local Edits: Embedding-Virtualized Knowledge for Broader Evaluation and Preservation of Model Editing

  2026influential citation
• Document-Level Zero-Shot Relation Extraction with Entity Side Information

  2026cites this paper
• Are We Evaluating the Edit Locality of LLM Model Editing Properly?

  2026influential citation
• G-MemLLM: Gated Latent Memory Augmentation for Long-Context Reasoning in Large Language Models

  2026influential citation
• Multiplicative Orthogonal Sequential Editing for Language Models

  2026cites this paper
• CLM-Bench: Benchmarking and Analyzing Cross-lingual Misalignment of LLMs in Knowledge Editing

  2026influential citation
• MoEEdit: Efficient and Routing-Stable Knowledge Editing for Mixture-of-Experts LLMs

  2026cites this paper
• Hierarchical Orthogonal Residual Spread for Precise Massive Editing in Large Language Models

  2026cites this paper
• Toward Ultra-Long-Horizon Sequential Model Editing

  2026influential citation
• Golden Layers and Where to Find Them: Improved Knowledge Editing for Large Language Models Via Layer Gradient Analysis

  2026cites this paper
• CACRE: A Weakly Supervised Method for Cross-Attention Contrastive Relation Extraction With Large Language Models

  2026cites this paper
• A tool-augmented LLM workflow for automated research contribution patterns extraction from citation sentences

  2026cites this paper
• Prompt-Contrastive Learning for Zero-Shot Relation Extraction

  2026influential citation
• SocraticKG: Knowledge Graph Construction via QA-Driven Fact Extraction

  2026cites this paper
• Autoencoding-Free Context Compression for LLMs via Contextual Semantic Anchors

  2025cites this paper
• ZS-HPDG: Hybrid Prompt-Tuning and Generation Data for Better Zero-Shot Relation Extraction

  2025cites this paper
• ZeroGR: A Generalizable and Scalable Framework for Zero-Shot Generative Retrieval

  2025cites this paper
• EMSEdit: Efficient Multi-Step Meta-Learning-based Model Editing

  2025cites this paper
• A Survey on Generative Model Unlearning: Fundamentals, Taxonomy, Evaluation, and Future Direction

  2025cites this paper
• Fine-tuning Done Right in Model Editing

  2025influential citation
• Muon Outperforms Adam in Tail-End Associative Memory Learning

  2025cites this paper
• Is Model Editing Built on Sand? Revealing Its Illusory Success and Fragile Foundation

  2025cites this paper
• To model human linguistic prediction, make LLMs less superhuman

  2025cites this paper
• Comparing Knowledge Injection Methods for LLMs in a Low-Resource Regime

  2025cites this paper
• Edited Knowledge Fusion with Large Language Models

  2025cites this paper
• Parameter-Aware Contrastive Knowledge Editing: Tracing and Rectifying based on Critical Transmission Paths

  2025cites this paper
• Event Extraction Based on Prompt Learning

  2025cites this paper
• A Dual-Axis Taxonomy of Knowledge Editing for LLMs: From Mechanisms to Functions

  2025influential citation
• Large Language Model Sourcing: A Survey

  2025cites this paper
• Zero-Shot Open-Schema Entity Structure Discovery

  2025cites this paper
• GuideX: Guided Synthetic Data Generation for Zero-Shot Information Extraction

  2025cites this paper
• MEMOIR: Lifelong Model Editing with Minimal Overwrite and Informed Retention for LLMs

  2025cites this paper
• Erasing Without Remembering: Implicit Knowledge Forgetting in Large Language Models

  2025cites this paper
• On the Superimposed Noise Accumulation Problem in Sequential Knowledge Editing of Large Language Models

  2025cites this paper
• Enhancing Language Model Hypernetworks with Restart: A Study on Optimization

  2025cites this paper
• InComeS: Integrating Compression and Selection Mechanisms into LLMs for Efficient Model Editing

  2025influential citation
• Survey on Factuality in Large Language Models

  2025cites this paper
• ImpRAG: Retrieval-Augmented Generation with Implicit Queries

  2025cites this paper
• Beyond Memorization: A Rigorous Evaluation Framework for Medical Knowledge Editing

  2025cites this paper
• MobiEdit: Resource-efficient Knowledge Editing for Personalized On-device LLMs

  2025cites this paper
• NeuralDB: Scaling Knowledge Editing in LLMs to 100,000 Facts with Neural KV Database

  2025cites this paper
• LoRA-Augmented Generation (LAG) for Knowledge-Intensive Language Tasks

  2025cites this paper
• Towards a Principled Evaluation of Knowledge Editors

  2025cites this paper
• An Exploration of Knowledge Editing for Arabic

  2025cites this paper
• Prefix-tuning-based Consistency Learning Framework for Machine Reading Comprehension with unanswerable questions

  2025cites this paper
• Aligning Language Models with Real-time Knowledge Editing

  2025cites this paper
• Latent Knowledge Scalpel: Precise and Massive Knowledge Editing for Large Language Models

  2025cites this paper
• Layerwise Importance Analysis of Feed-Forward Networks in Transformer-based Language Models

  2025cites this paper
• From Construction to Injection: Edit-Based Fingerprints for Large Language Models

  2025cites this paper
• UltraEdit: Training-, Subject-, and Memory-Free Lifelong Editing in Language Models

  2025cites this paper
• Knowledge Editing with Subspace-Aware Key-Value Mappings

  2025cites this paper
• Knowledge Acquisition Method of Urban Rail Transit Safety Event Case Base for Intelligent Emergency Response

  2025cites this paper
• Energy-Regularized Sequential Model Editing on Hyperspheres

  2025influential citation
• REPAIR: Robust Editing via Progressive Adaptive Intervention and Reintegration

  2025cites this paper
• KnowledgeSmith: Uncovering Knowledge Updating in LLMs with Model Editing and Unlearning

  2025cites this paper
• Memorization and Knowledge Injection in Gated LLMs

  2025cites this paper
• Assessing and Post-Processing Black Box Large Language Models for Knowledge Editing

  2025cites this paper
• Generating Complex Question Decompositions in the Face of Distribution Shifts

  2025cites this paper
• LyapLock: Bounded Knowledge Preservation in Sequential Large Language Model Editing

  2025influential citation
• PropMEND: Hypernetworks for Knowledge Propagation in LLMs

  2025cites this paper
• Training a Utility-based Retriever Through Shared Context Attribution for Retrieval-Augmented Language Models

  2025cites this paper
• A Survey on Knowledge-Oriented Retrieval-Augmented Generation

  2025cites this paper
• Efficient Tuning of Large Language Models for Knowledge-Grounded Dialogue Generation

  2025cites this paper
• Constraining Sequential Model Editing with Editing Anchor Compression

  2025influential citation
• Capability Localization: Capabilities Can be Localized rather than Individual Knowledge

  2025cites this paper
• MindBridge: Scalable and Cross-Model Knowledge Editing via Memory-Augmented Modality

  2025cites this paper
• GeoEdit: Geometric Knowledge Editing for Large Language Models

  2025influential citation
• WikiBigEdit: Understanding the Limits of Lifelong Knowledge Editing in LLMs

  2025cites this paper
• The Missing Piece in Model Editing: A Deep Dive into the Hidden Damage Brought By Model Editing

  2025cites this paper
• MARRO: Multi-headed Attention for Rhetorical Role Labeling in Legal Documents

  2025cites this paper
• Resolving UnderEdit & OverEdit with Iterative & Neighbor-Assisted Model Editing

  2025influential citation
• Explainable and Efficient Editing for Large Language Models

  2025cites this paper
• Decoupling Knowledge and Context: An Efficient and Effective Retrieval Augmented Generation Framework via Cross Attention

  2025cites this paper
• How Effective are Generative Large Language Models in Performing Requirements Classification?

  2025cites this paper
• SetKE: Knowledge Editing for Knowledge Elements Overlap

  2025cites this paper
• SDZRE: A semantic distillation method for zero-shot relation extraction

  2025cites this paper
• Osiris: A Lightweight Open-Source Hallucination Detection System

  2025cites this paper
• PT-MoE: An Efficient Finetuning Framework for Integrating Mixture-of-Experts into Prompt Tuning

  2025cites this paper
• EAMET: Robust Massive Model Editing via Embedding Alignment Optimization

  2025cites this paper
• DocMEdit: Towards Document-Level Model Editing

  2025cites this paper
• Benchmarking and Rethinking Knowledge Editing for Large Language Models

  2025cites this paper
• Context-Robust Knowledge Editing for Language Models

  2025cites this paper
• Boosting Retrieval-Augmented Generation with Generation-Augmented Retrieval: A Co-Training Approach

  2025cites this paper
• COMPKE: Complex Question Answering under Knowledge Editing

  2025cites this paper
• One for All: Update Parameterized Knowledge Across Multiple Models

  2025influential citation
• KARE-RAG: Knowledge-Aware Refinement and Enhancement for RAG

  2025cites this paper
• Unveiling and Eliminating the Shortcut Learning for Locate-Then-Edit Knowledge Editing via Both Subject and Relation Awareness

  2025cites this paper
• Novel Sequential Knowledge Batch Editing via Retrieving Padding Right Embedding

  2025cites this paper
• Model Merging for Knowledge Editing

  2025cites this paper
• Instruction Tuning with and without Context: Behavioral Shifts and Downstream Impact

  2025cites this paper
• QueueEDIT: Structural Self-Correction for Sequential Model Editing in LLMs

  2025cites this paper
• AdaEdit: Advancing Continuous Knowledge Editing For Large Language Models

  2025cites this paper
• CoME: An Unlearning-based Approach to Conflict-free Model Editing

  2025influential citation