Recovering Question Answering Errors via Query Revision

The existing factoid QA systems often lack a post-inspection component that can help models recover from their own mistakes. In this work, we propose to crosscheck the corresponding KB relations behind the predicted answers and identify potential inconsistencies. Instead of developing a new model that accepts evidences collected from these relations, we choose to plug them back to the original questions directly and check if the revised question makes sense or not. A bidirectional LSTM is applied to encode revised questions. We develop a scoring mechanism over the revised question encodings to refine the predictions of a base QA system. This approach can improve the F1 score of STAGG (Yih et al., 2015), one of the leading QA systems, from 52.5% to 53.9% on WEBQUESTIONS data.

• Publication year

  2017

• Venue

  Conference on Empirical Methods in Natural Language Processing

• Publication date

  2017-09-01

• Fields of study

  Computer Science

• Identifiers
  DOI 10.18653/v1/D17-1094
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Improving Semantic Parsing via Answer Type Inference

  2016influential reference
• Natural Language Comprehension with the EpiReader

  2016cited by this paper
• A Comparison of Deep Learning Based Query Expansion with Pseudo-Relevance Feedback and Mutual Information

  2016cited by this paper
• Query Expansion with Locally-Trained Word Embeddings

  2016cited by this paper
• TensorFlow: A system for large-scale machine learning

  2016cited by this paper
• A Thorough Examination of the CNN/Daily Mail Reading Comprehension Task

  2016cited by this paper
• Siamese Recurrent Architectures for Learning Sentence Similarity

  2016cited by this paper
• Learning Text Similarity with Siamese Recurrent Networks

  2016cited by this paper
• Question Answering on Freebase via Relation Extraction and Textual Evidence

  2016influential reference
• Hybrid Question Answering over Knowledge Base and Free Text

  2016influential reference
• Improved Representation Learning for Question Answer Matching

  2016influential reference
• Transforming Dependency Structures to Logical Forms for Semantic Parsing

  2016influential reference
• Large-scale Simple Question Answering with Memory Networks

  2015influential reference
• Semantic Parsing via Staged Query Graph Generation: Question Answering with Knowledge Base

  2015influential reference
• Scalable Semantic Parsing with Partial Ontologies

  2015cited by this paper
• Lean Question Answering over Freebase from Scratch

  2015influential reference
• Question Answering over Freebase with Multi-Column Convolutional Neural Networks

  2015influential reference
• Imitation Learning of Agenda-based Semantic Parsers

  2015influential reference
• Teaching Machines to Read and Comprehend

  2015cited by this paper
• Effective Approaches to Attention-based Neural Machine Translation

  2015cited by this paper
• Question Answering with Subgraph Embeddings

  2014influential reference
• Semantic Parsing via Paraphrasing

  2014influential reference
• Knowledge-Based Question Answering as Machine Translation

  2014influential reference
• GloVe: Global Vectors for Word Representation

  2014cited by this paper
• Information Extraction over Structured Data: Question Answering with Freebase

  2014influential reference
• Adam: A Method for Stochastic Optimization

  2014cited by this paper
• Learning Concept Embeddings for Query Expansion by Quantum Entropy Minimization

  2014cited by this paper
• The Stanford CoreNLP Natural Language Processing Toolkit

  2014cited by this paper
• Joint Relational Embeddings for Knowledge-based Question Answering

  2014influential reference
• Large-scale Semantic Parsing without Question-Answer Pairs

  2014cited by this paper
• Scaling Semantic Parsers with On-the-Fly Ontology Matching

  2013cited by this paper
• Concept Based Query Expansion

  2013cited by this paper
• Semantic Parsing on Freebase from Question-Answer Pairs

  2013influential reference
• Minimum Bayes Risk based Answer Re-ranking for Question Answering

  2013cited by this paper
• Freebase: a collaboratively created graph database for structuring human knowledge

  2008cited by this paper
• Improving Question Answering by Combining Multiple Systems Via Answer Validation

  2008cited by this paper
• A Re-examination of Query Expansion Using Lexical Resources

  2008cited by this paper
• Statistical Machine Translation for Query Expansion in Answer Retrieval

  2007cited by this paper
• DBpedia: A Nucleus for a Web of Open Data

  2007cited by this paper
• YAGO: A Core of Semantic Knowledge Unifying WordNet and Wikipedia

  2007cited by this paper
• An analysis of ontology-based query expansion strategies

  2003cited by this paper
• Improving automatic query expansion

  1998cited by this paper
• Concept based query expansion

  1993cited by this paper

• Bayesian-error-informed contrastive learning for knowledge-based question answering systems

  2025cites this paper
• Enhancing knowledge-based question answering with dense passage retrieval-based scaled momentum contrastive learning

  2025cites this paper
• A reasoning enhance network for muti-relation question answering

  2021cites this paper
• RNG-KBQA: Generation Augmented Iterative Ranking for Knowledge Base Question Answering

  2021cites this paper
• Modelling Long-distance Node Relations for KBQA with Global Dynamic Graph

  2020cites this paper
• Towards Democratizing Data Science with Natural Language Interfaces

  2019cites this paper
• DeepAssist: Deep Knowledge Grounding for Factual and Conversational Natural Language Interfaces

  2019cites this paper
• An Interpretable Reasoning Network for Multi-Relation Question Answering

  2018cites this paper