Bidirectional Attention Flow for Machine Comprehension

Abstract

Machine comprehension (MC), answering a query about a given context paragraph, requires modeling complex interactions between the context and the query. Recently, attention mechanisms have been successfully extended to MC. Typically these methods use attention to focus on a small portion of the context and summarize it with a fixed-size vector, couple attentions temporally, and/or often form a uni-directional attention. In this paper we introduce the Bi-Directional Attention Flow (BIDAF) network, a multi-stage hierarchical process that represents the context at different levels of granularity and uses bi-directional attention flow mechanism to obtain a query-aware context representation without early summarization. Our experimental evaluations show that our model achieves the state-of-the-art results in Stanford Question Answering Dataset (SQuAD) and CNN/DailyMail cloze test.

Published as a conference paper at ICLR 2017

Cited by in corpus (363)

• A General Survey on Attention Mechanisms in Deep Learning
• GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
• REALM: Retrieval-Augmented Language Model Pre-Training
• Passage Re-ranking with BERT
• Enhancing Clinical Concept Extraction with Contextual Embeddings
• CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge
• ReasoNet: Learning to Stop Reading in Machine Comprehension
• Adversarial Attacks on Deep Learning Models in Natural Language Processing: A Survey
• A Survey of Knowledge-Enhanced Text Generation
• LXMERT: Learning Cross-Modality Encoder Representations from Transformers
• ReCoRD: Bridging the Gap between Human and Machine Commonsense Reading Comprehension
• BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions
• Spatial Group-wise Enhance: Improving Semantic Feature Learning in Convolutional Networks
• ST-GRAT: A Novel Spatio-temporal Graph Attention Network for Accurately Forecasting Dynamically Changing Road Speed
• Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answering
• Learning to Retrieve Reasoning Paths over Wikipedia Graph for Question Answering
• K-Adapter: Infusing Knowledge into Pre-Trained Models with Adapters
• Towards Automated ICD Coding Using Deep Learning
• Biomedical Question Answering: A Survey of Approaches and Challenges
• DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs
• Natural Language Video Localization: A Revisit in Span-based Question Answering Framework
• It's Morphin' Time! Combating Linguistic Discrimination with Inflectional Perturbations
• CharBERT: Character-aware Pre-trained Language Model
• Reading Wikipedia to Answer Open-Domain Questions
• Paradigm Shift in Natural Language Processing
• Exploring Graph-structured Passage Representation for Multi-hop Reading Comprehension with Graph Neural Networks
• Representation Learning for Natural Language Processing
• Neural Reverse Engineering of Stripped Binaries using Augmented Control Flow Graphs
• Focal Visual-Text Attention for Visual Question Answering
• Temporal Sentence Grounding in Videos: A Survey and Future Directions
• Graph Neural Networks for Natural Language Processing: A Survey
• Representation Learning for Dynamic Graphs: A Survey
• CAN-NER: Convolutional Attention Network for Chinese Named Entity Recognition
• Multi-hop Question Answering via Reasoning Chains
• Deep Neural Approaches to Relation Triplets Extraction: A Comprehensive Survey
• Multi-domain Dialogue State Tracking as Dynamic Knowledge Graph Enhanced Question Answering
• Joint Extraction of Entities and Relations Based on a Novel Decomposition Strategy
• Learning Dynamic Belief Graphs to Generalize on Text-Based Games
• Enhancing Knowledge Retrieval with In-Context Learning and Semantic Search through Generative AI
• A Unified MRC Framework for Named Entity Recognition
• A Unified Query-based Generative Model for Question Generation and Question Answering
• Coarse-grain Fine-grain Coattention Network for Multi-evidence Question Answering
• Entity-Relation Extraction as Multi-Turn Question Answering
• Machine Reading Comprehension: The Role of Contextualized Language Models and Beyond
• DUMA: Reading Comprehension with Transposition Thinking
• Cross-Lingual Machine Reading Comprehension
• Challenges in Building Intelligent Open-domain Dialog Systems
• Differentiable Reasoning over a Virtual Knowledge Base
• Dynamically Fused Graph Network for Multi-hop Reasoning
• Parallel Attention Network with Sequence Matching for Video Grounding
• Enhancing textual textbook question answering with large language models and retrieval augmented generation
• Multi-hop Question Generation with Graph Convolutional Network
• Is Graph Structure Necessary for Multi-hop Question Answering?
• U-Net: Machine Reading Comprehension with Unanswerable Questions
• Conversational Question Reformulation via Sequence-to-Sequence Architectures and Pretrained Language Models
• LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attention
• Stochastic Answer Networks for Machine Reading Comprehension
• Enhancing lexical-based approach with external knowledge for Vietnamese multiple-choice machine reading comprehension
• Multi-range Reasoning for Machine Comprehension
• Dice Loss for Data-imbalanced NLP Tasks
• Description Based Text Classification with Reinforcement Learning
• Retrieve-and-Read: Multi-task Learning of Information Retrieval and Reading Comprehension
• Putting Question-Answering Systems into Practice: Transfer Learning for Efficient Domain Customization
• Convolutional Spatial Attention Model for Reading Comprehension with Multiple-Choice Questions
• Hierarchical Graph Network for Multi-hop Question Answering
• Neural Machine Reading Comprehension: Methods and Trends
• SpeechBERT: An Audio-and-text Jointly Learned Language Model for End-to-end Spoken Question Answering
• Semi-Supervised QA with Generative Domain-Adaptive Nets
• Don't Take the Easy Way Out: Ensemble Based Methods for Avoiding Known Dataset Biases
• Cognitive Graph for Multi-Hop Reading Comprehension at Scale
• Spoken SQuAD: A Study of Mitigating the Impact of Speech Recognition Errors on Listening Comprehension
• Towards a Robust Deep Neural Network in Texts: A Survey
• Enhance word representation for out-of-vocabulary on Ubuntu dialogue corpus
• On the Effectiveness of Low-Rank Matrix Factorization for LSTM Model Compression
• SG-Net: Syntax-Guided Machine Reading Comprehension
• TVQA+: Spatio-Temporal Grounding for Video Question Answering
• Zero-shot Reading Comprehension by Cross-lingual Transfer Learning with Multi-lingual Language Representation Model
• Multi-passage BERT: A Globally Normalized BERT Model for Open-domain Question Answering
• From 'F' to 'A' on the N.Y. Regents Science Exams: An Overview of the Aristo Project
• Event Extraction by Answering (Almost) Natural Questions
• SUMBT: Slot-Utterance Matching for Universal and Scalable Belief Tracking
• Sequence tagging for biomedical extractive question answering
• Open-Domain Targeted Sentiment Analysis via Span-Based Extraction and Classification
• Dialog State Tracking: A Neural Reading Comprehension Approach
• KG^2: Learning to Reason Science Exam Questions with Contextual Knowledge Graph Embeddings
• Multi-Cast Attention Networks for Retrieval-based Question Answering and Response Prediction
• Themis: Fair and Efficient GPU Cluster Scheduling
• Logiformer: A Two-Branch Graph Transformer Network for Interpretable Logical Reasoning
• QA4IE: A Question Answering based Framework for Information Extraction
• Dynamic Fusion Networks for Machine Reading Comprehension
• Show Your Work: Improved Reporting of Experimental Results
• Coreferential Reasoning Learning for Language Representation
• MKD: a Multi-Task Knowledge Distillation Approach for Pretrained Language Models
• Reasoning with Sarcasm by Reading In-between
• Stochastic Answer Networks for SQuAD 2.0
• Conversing by Reading: Contentful Neural Conversation with On-demand Machine Reading
• TVR: A Large-Scale Dataset for Video-Subtitle Moment Retrieval
• EDUQA: Educational Domain Question Answering System using Conceptual Network Mapping
• Question Answering through Transfer Learning from Large Fine-grained Supervision Data
• To compress or not to compress: Understanding the Interactions between Adversarial Attacks and Neural Network Compression
• Improving Background Based Conversation with Context-aware Knowledge Pre-selection
• Discriminative Sentence Modeling for Story Ending Prediction
• Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index
• Compositional Questions Do Not Necessitate Multi-hop Reasoning
• Dynaboard: An Evaluation-As-A-Service Platform for Holistic Next-Generation Benchmarking
• A Discrete Hard EM Approach for Weakly Supervised Question Answering
• Improving Natural Language Inference Using External Knowledge in the Science Questions Domain
• Difficulty Controllable Generation of Reading Comprehension Questions
• Unsupervised Domain Adaptation of Language Models for Reading Comprehension
• Span-based Localizing Network for Natural Language Video Localization
• Knowledge Graph-based Question Answering with Electronic Health Records
• Question Answering over Knowledge Base using Language Model Embeddings
• Finding Answers from the Word of God: Domain Adaptation for Neural Networks in Biblical Question Answering
• ELI5: Long Form Question Answering
• RikiNet: Reading Wikipedia Pages for Natural Question Answering
• AllenNLP Interpret: A Framework for Explaining Predictions of NLP Models
• Towards Domain Adaptation from Limited Data for Question Answering Using Deep Neural Networks
• Ruminating Reader: Reasoning with Gated Multi-Hop Attention
• Commonsense for Generative Multi-Hop Question Answering Tasks
• Multi-hop Reading Comprehension through Question Decomposition and Rescoring
• Ranking Paragraphs for Improving Answer Recall in Open-Domain Question Answering
• ERICA: Improving Entity and Relation Understanding for Pre-trained Language Models via Contrastive Learning
• Machine Reading Comprehension for Answer Re-Ranking in Customer Support Chatbots
• Learning Contextualized Document Representations for Healthcare Answer Retrieval
• Answering Science Exam Questions Using Query Rewriting with Background Knowledge
• Multi-style Generative Reading Comprehension
• ExpMRC: Explainability Evaluation for Machine Reading Comprehension
• VLSP 2021 - ViMRC Challenge: Vietnamese Machine Reading Comprehension
• Graph-free Multi-hop Reading Comprehension: A Select-to-Guide Strategy
• Multi-tasking Dialogue Comprehension with Discourse Parsing
• Parameter-Efficient Abstractive Question Answering over Tables or Text
• Interpretable Multi-Step Reasoning with Knowledge Extraction on Complex Healthcare Question Answering
• Multi-Passage Machine Reading Comprehension with Cross-Passage Answer Verification
• Towards Exploiting Background Knowledge for Building Conversation Systems
• Knowledge Aware Conversation Generation with Explainable Reasoning over Augmented Graphs
• Gated Embeddings in End-to-End Speech Recognition for Conversational-Context Fusion
• Addressing Semantic Drift in Question Generation for Semi-Supervised Question Answering
• DramaQA: Character-Centered Video Story Understanding with Hierarchical QA
• Legal Element-oriented Modeling with Multi-view Contrastive Learning for Legal Case Retrieval
• DC-BERT: Decoupling Question and Document for Efficient Contextual Encoding
• Can You Tell Me How to Get Past Sesame Street? Sentence-Level Pretraining Beyond Language Modeling
• A Survey of Document Grounded Dialogue Systems (DGDS)
• CHOPT : Automated Hyperparameter Optimization Framework for Cloud-Based Machine Learning Platforms
• VIOLIN: A Large-Scale Dataset for Video-and-Language Inference
• Explore, Propose, and Assemble: An Interpretable Model for Multi-Hop Reading Comprehension
• A Survey on Machine Reading Comprehension: Tasks, Evaluation Metrics and Benchmark Datasets
• Understanding Dataset Design Choices for Multi-hop Reasoning
• TextFlint: Unified Multilingual Robustness Evaluation Toolkit for Natural Language Processing
• Towards Generalizable Neuro-Symbolic Systems for Commonsense Question Answering
• Explicit Utilization of General Knowledge in Machine Reading Comprehension
• Multi-Stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting
• Extracting Symptoms and their Status from Clinical Conversations
• A Qualitative Comparison of CoQA, SQuAD 2.0 and QuAC
• RECONSIDER: Re-Ranking using Span-Focused Cross-Attention for Open Domain Question Answering
• Dual Pointer Network for Fast Extraction of Multiple Relations in a Sentence
• Learning to Coordinate Multiple Reinforcement Learning Agents for Diverse Query Reformulation
• Controlling Risk of Web Question Answering
• Improving Machine Reading Comprehension via Adversarial Training
• Classification of Goods Using Text Descriptions With Sentences Retrieval
• Cosmos QA: Machine Reading Comprehension with Contextual Commonsense Reasoning
• QAInfomax: Learning Robust Question Answering System by Mutual Information Maximization
• Thinking Globally, Acting Locally: Distantly Supervised Global-to-Local Knowledge Selection for Background Based Conversation
• A Multi-Type Multi-Span Network for Reading Comprehension that Requires Discrete Reasoning
• A Vietnamese Dataset for Evaluating Machine Reading Comprehension
• SRQA: Synthetic Reader for Factoid Question Answering
• RefNet: A Reference-aware Network for Background Based Conversation
• Answering Questions on COVID-19 in Real-Time
• Adversarial Domain Adaptation for Machine Reading Comprehension
• Benchmarking Knowledge-Enhanced Commonsense Question Answering via Knowledge-to-Text Transformation
• Sanity Check: A Strong Alignment and Information Retrieval Baseline for Question Answering
• Enhancing Answer Boundary Detection for Multilingual Machine Reading Comprehension
• Enriching a Model's Notion of Belief using a Persistent Memory
• Exploring BERT Parameter Efficiency on the Stanford Question Answering Dataset v2.0
• Stress Test Evaluation of Transformer-based Models in Natural Language Understanding Tasks
• Multi-task Learning with Sample Re-weighting for Machine Reading Comprehension
• Document Modeling with Graph Attention Networks for Multi-grained Machine Reading Comprehension
• Sogou Machine Reading Comprehension Toolkit
• Incorporating External Knowledge into Machine Reading for Generative Question Answering
• Temporal Sub-sampling of Audio Feature Sequences for Automated Audio Captioning
• A Framework for Evaluation of Machine Reading Comprehension Gold Standards
• Conversations with Search Engines: SERP-based Conversational Response Generation
• Large-scale Multi-granular Concept Extraction Based on Machine Reading Comprehension
• Answering while Summarizing: Multi-task Learning for Multi-hop QA with Evidence Extraction
• Improving Robustness and Generality of NLP Models Using Disentangled Representations
• Multi-span Style Extraction for Generative Reading Comprehension
• BERTQA -- Attention on Steroids
• Multi-Task Learning for Conversational Question Answering over a Large-Scale Knowledge Base
• Tensorized Self-Attention: Efficiently Modeling Pairwise and Global Dependencies Together
• Can NLI Models Verify QA Systems' Predictions?
• Bridging the Gap between Language Model and Reading Comprehension: Unsupervised MRC via Self-Supervision
• Why Machine Reading Comprehension Models Learn Shortcuts?
• Act-Aware Slot-Value Predicting in Multi-Domain Dialogue State Tracking
• Weakly Supervised Reasoning by Neuro-Symbolic Approaches
• Co-attention Graph Pooling for Efficient Pairwise Graph Interaction Learning
• An MRC Framework for Semantic Role Labeling
• EfficientQA : a RoBERTa Based Phrase-Indexed Question-Answering System
• Diverse, Controllable, and Keyphrase-Aware: A Corpus and Method for News Multi-Headline Generation
• A Self-Training Method for Machine Reading Comprehension with Soft Evidence Extraction
• T3: Tree-Autoencoder Constrained Adversarial Text Generation for Targeted Attack
• Improving Generalization by Incorporating Coverage in Natural Language Inference
• Self-Assembling Modular Networks for Interpretable Multi-Hop Reasoning
• LATTE: Latent Type Modeling for Biomedical Entity Linking
• ChID: A Large-scale Chinese IDiom Dataset for Cloze Test
• Episodic Memory Reader: Learning What to Remember for Question Answering from Streaming Data
• FastFusionNet: New State-of-the-Art for DAWNBench SQuAD
• Fast Prototyping a Dialogue Comprehension System for Nurse-Patient Conversations on Symptom Monitoring
• Gaining Extra Supervision via Multi-task learning for Multi-Modal Video Question Answering
• Joint Effects of Context and User History for Predicting Online Conversation Re-entries
• Learning to Ask Unanswerable Questions for Machine Reading Comprehension
• Avoiding Reasoning Shortcuts: Adversarial Evaluation, Training, and Model Development for Multi-Hop QA
• Query-by-example Spoken Term Detection using Attention-based Multi-hop Networks
• Contextual embedding and model weighting by fusing domain knowledge on Biomedical Question Answering
• Adaptive Bi-directional Attention: Exploring Multi-Granularity Representations for Machine Reading Comprehension
• Compressing LSTM Networks by Matrix Product Operators
• NumNet: Machine Reading Comprehension with Numerical Reasoning
• Towards Socially Intelligent Agents with Mental State Transition and Human Utility
• JEC-QA: A Legal-Domain Question Answering Dataset
• Lightweight Convolutional Approaches to Reading Comprehension on SQuAD
• Observing Dialogue in Therapy: Categorizing and Forecasting Behavioral Codes
• Interventional Video Grounding with Dual Contrastive Learning
• Careful Selection of Knowledge to solve Open Book Question Answering
• Effective Subword Segmentation for Text Comprehension
• Conversational Question Answering: A Survey
• Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?
• Harvesting and Refining Question-Answer Pairs for Unsupervised QA
• DSReg: Using Distant Supervision as a Regularizer
• Large-scale Cloze Test Dataset Created by Teachers
• Make Text Unlearnable: Exploiting Effective Patterns to Protect Personal Data
• NeuronBlocks: Building Your NLP DNN Models Like Playing Lego
• Adaptations of ROUGE and BLEU to Better Evaluate Machine Reading Comprehension Task
• Integrated Triaging for Fast Reading Comprehension
• AntMan: Sparse Low-Rank Compression to Accelerate RNN inference
• Zero-shot Text-to-SQL Learning with Auxiliary Task
• Multi-Mention Learning for Reading Comprehension with Neural Cascades
• Deep Neural Networks for Relation Extraction
• CliniQG4QA: Generating Diverse Questions for Domain Adaptation of Clinical Question Answering
• "Bilingual Expert" Can Find Translation Errors
• Evaluating Neural Machine Comprehension Model Robustness to Noisy Inputs and Adversarial Attacks
• Look at the First Sentence: Position Bias in Question Answering
• Probing What Different NLP Tasks Teach Machines about Function Word Comprehension
• Noisy Text Data: Achilles' Heel of popular transformer based NLP models
• Reinforced Dynamic Reasoning for Conversational Question Generation
• TWEETQA: A Social Media Focused Question Answering Dataset
• Stay Hungry, Stay Focused: Generating Informative and Specific Questions in Information-Seeking Conversations
• ListReader: Extracting List-form Answers for Opinion Questions
• Clinical Reading Comprehension: A Thorough Analysis of the emrQA Dataset
• R3: A Reading Comprehension Benchmark Requiring Reasoning Processes
• Sequential Attention Module for Natural Language Processing
• Coarse-grained decomposition and fine-grained interaction for multi-hop question answering
• Low Rank Factorization for Compact Multi-Head Self-Attention
• EQuANt (Enhanced Question Answer Network)
• AnswerQuest: A System for Generating Question-Answer Items from Multi-Paragraph Documents
• Composing Answer from Multi-spans for Reading Comprehension
• Keyword-based Query Comprehending via Multiple Optimized-Demand Augmentation
• Deep Understanding based Multi-Document Machine Reading Comprehension
• Trying Bilinear Pooling in Video-QA
• FAQ-based Question Answering via Knowledge Anchors
• Textbook Question Answering with Multi-modal Context Graph Understanding and Self-supervised Open-set Comprehension
• Teaching Machine Comprehension with Compositional Explanations
• Question Directed Graph Attention Network for Numerical Reasoning over Text
• PolicyQA: A Reading Comprehension Dataset for Privacy Policies
• Answering Open-Domain Questions of Varying Reasoning Steps from Text
• BiSET: Bi-directional Selective Encoding with Template for Abstractive Summarization
• Exploring Fluent Query Reformulations with Text-to-Text Transformers and Reinforcement Learning
• GenNet : Reading Comprehension with Multiple Choice Questions using Generation and Selection model
• A Simple but Effective Method to Incorporate Multi-turn Context with BERT for Conversational Machine Comprehension
• The Geometry of Distributed Representations for Better Alignment, Attenuated Bias, and Improved Interpretability
• Non-parametric Structural Change Detection in Multivariate Systems
• Discriminative Pre-training for Low Resource Title Compression in Conversational Grocery
• Microblog Hashtag Generation via Encoding Conversation Contexts
• MS-Ranker: Accumulating Evidence from Potentially Correct Candidates for Answer Selection
• Relation Module for Non-answerable Prediction on Question Answering
• Subjective Question Answering: Deciphering the inner workings of Transformers in the realm of subjectivity
• Are Multilingual BERT models robust? A Case Study on Adversarial Attacks for Multilingual Question Answering
• Large Scale Question Answering using Tourism Data
• An Empirical Study of Content Understanding in Conversational Question Answering
• Dense-Caption Matching and Frame-Selection Gating for Temporal Localization in VideoQA
• Let's Ask Again: Refine Network for Automatic Question Generation
• TMT: A Transformer-based Modal Translator for Improving Multimodal Sequence Representations in Audio Visual Scene-aware Dialog
• XCMRC: Evaluating Cross-lingual Machine Reading Comprehension
• DuReader_robust: A Chinese Dataset Towards Evaluating Robustness and Generalization of Machine Reading Comprehension in Real-World Applications
• Syntax-Enhanced Pre-trained Model
• Correction of Faulty Background Knowledge based on Condition Aware and Revise Transformer for Question Answering
• Attend to the beginning: A study on using bidirectional attention for extractive summarization
• MCR-Net: A Multi-Step Co-Interactive Relation Network for Unanswerable Questions on Machine Reading Comprehension
• Tradeoffs in Sentence Selection Techniques for Open-Domain Question Answering
• A Compare Aggregate Transformer for Understanding Document-grounded Dialogue
• Situation and Behavior Understanding by Trope Detection on Films
• BiPaR: A Bilingual Parallel Dataset for Multilingual and Cross-lingual Reading Comprehension on Novels
• What's Missing: A Knowledge Gap Guided Approach for Multi-hop Question Answering
• A Split-and-Recombine Approach for Follow-up Query Analysis
• Everything Happens for a Reason: Discovering the Purpose of Actions in Procedural Text
• Go Beyond Plain Fine-tuning: Improving Pretrained Models for Social Commonsense
• BERT-CoQAC: BERT-based Conversational Question Answering in Context
• QuASE: Question-Answer Driven Sentence Encoding
• Simple yet Effective Bridge Reasoning for Open-Domain Multi-Hop Question Answering
• How Optimal is Greedy Decoding for Extractive Question Answering?
• Learning Representations and Agents for Information Retrieval
• Propagate-Selector: Detecting Supporting Sentences for Question Answering via Graph Neural Networks
• Zero-Shot Open-Book Question Answering
• Query-Based Named Entity Recognition
• Learning Question-Guided Video Representation for Multi-Turn Video Question Answering
• Katecheo: A Portable and Modular System for Multi-Topic Question Answering
• Attending to Entities for Better Text Understanding
• An Understanding-Oriented Robust Machine Reading Comprehension Model
• Continual Machine Reading Comprehension via Uncertainty-aware Fixed Memory and Adversarial Domain Adaptation
• A Simple Recurrent Unit with Reduced Tensor Product Representations
• Interactive Machine Comprehension with Dynamic Knowledge Graphs
• Pay More Attention - Neural Architectures for Question-Answering
• Knowledge Distillation for Improved Accuracy in Spoken Question Answering
• Improving QA Generalization by Concurrent Modeling of Multiple Biases
• Incremental Reading for Question Answering
• Multi-Perspective Fusion Network for Commonsense Reading Comprehension
• Knowing More About Questions Can Help: Improving Calibration in Question Answering
• Extracting Factual Min/Max Age Information from Clinical Trial Studies
• Practical Annotation Strategies for Question Answering Datasets
• Machine Reading Comprehension using Case-based Reasoning
• Effective Attention Modeling for Neural Relation Extraction
• Self-supervised Dialogue Learning for Spoken Conversational Question Answering
• Zero-Shot Estimation of Base Models' Weights in Ensemble of Machine Reading Comprehension Systems for Robust Generalization
• Bew: Towards Answering Business-Entity-Related Web Questions
• Bi-directional Cognitive Thinking Network for Machine Reading Comprehension
• Contrastive Domain Adaptation for Question Answering using Limited Text Corpora
• Improving Numerical Reasoning Skills in the Modular Approach for Complex Question Answering on Text
• Orthogonal Attention: A Cloze-Style Approach to Negation Scope Resolution
• Learning from Language Description: Low-shot Named Entity Recognition via Decomposed Framework
• Incorporating Connections Beyond Knowledge Embeddings: A Plug-and-Play Module to Enhance Commonsense Reasoning in Machine Reading Comprehension
• Ensembling Strategies for Answering Natural Questions
• Hierarchical Attention: What Really Counts in Various NLP Tasks
• ODSQA: Open-domain Spoken Question Answering Dataset
• Effective Character-augmented Word Embedding for Machine Reading Comprehension
• A Framework for Rationale Extraction for Deep QA models
• Reasoning-Driven Question-Answering for Natural Language Understanding
• QA4PRF: A Question Answering based Framework for Pseudo Relevance Feedback
• ClueReader: Heterogeneous Graph Attention Network for Multi-hop Machine Reading Comprehension
• Why can't memory networks read effectively?
• Multi-hop Reading Comprehension via Deep Reinforcement Learning based Document Traversal
• Two-stream Spatiotemporal Feature for Video QA Task
• Solving Price Per Unit Problem Around the World: Formulating Fact Extraction as Question Answering
• SRLGRN: Semantic Role Labeling Graph Reasoning Network
• Retrofitting Contextualized Word Embeddings with Paraphrases
• Incorporating Commonsense Knowledge Graph in Pretrained Models for Social Commonsense Tasks
• A Logic-Driven Framework for Consistency of Neural Models
• Hierarchical RNN with Static Sentence-Level Attention for Text-Based Speaker Change Detection
• AutoNLU: An On-demand Cloud-based Natural Language Understanding System for Enterprises
• Segmenting Natural Language Sentences via Lexical Unit Analysis
• Knowledge-guided Open Attribute Value Extraction with Reinforcement Learning
• When to Fold'em: How to answer Unanswerable questions
• Bilingual Text Extraction as Reading Comprehension
• CAESAR: Context Awareness Enabled Summary-Attentive Reader
• DCA: Diversified Co-Attention towards Informative Live Video Commenting
• MaP: A Matrix-based Prediction Approach to Improve Span Extraction in Machine Reading Comprehension
• MIX : a Multi-task Learning Approach to Solve Open-Domain Question Answering
• Differentiable Window for Dynamic Local Attention
• Response-Anticipated Memory for On-Demand Knowledge Integration in Response Generation
• Neighbourhood Distillation: On the benefits of non end-to-end distillation
• ComQA:Compositional Question Answering via Hierarchical Graph Neural Networks
• Do not let the history haunt you -- Mitigating Compounding Errors in Conversational Question Answering
• Muppet: Massive Multi-task Representations with Pre-Finetuning
• Delta Embedding Learning
• Weak Supervision Enhanced Generative Network for Question Generation
• Weighted Global Normalization for Multiple Choice Reading Comprehension over Long Documents
• Multi-Relational Question Answering from Narratives: Machine Reading and Reasoning in Simulated Worlds