SOTAVerified

Question Generation

The goal of Question Generation is to generate a valid and fluent question according to a given passage and the target answer. Question Generation can be used in many scenarios, such as automatic tutoring systems, improving the performance of Question Answering models and enabling chatbots to lead a conversation.

Source: Generating Highly Relevant Questions

Papers

Showing 76100 of 664 papers

TitleStatusHype
Cooperative Self-training of Machine Reading ComprehensionCode1
AnswerQuest: A System for Generating Question-Answer Items from Multi-Paragraph DocumentsCode1
Quiz-Style Question Generation for News StoriesCode1
ChainCQG: Flow-Aware Conversational Question GenerationCode1
BANG: Bridging Autoregressive and Non-autoregressive Generation with Large Scale PretrainingCode1
Contrastive Learning with Adversarial Perturbations for Conditional Text GenerationCode1
EQG-RACE: Examination-Type Question GenerationCode1
Just Ask: Learning to Answer Questions from Millions of Narrated VideosCode1
Exploring Question-Specific Rewards for Generating Deep QuestionsCode1
PathQG: Neural Question Generation from FactsCode1
CliniQG4QA: Generating Diverse Questions for Domain Adaptation of Clinical Question AnsweringCode1
Unsupervised Multi-hop Question Answering by Question GenerationCode1
Multi-hop Question Generation with Graph Convolutional NetworkCode1
Contrast and Classify: Training Robust VQA ModelsCode1
Mathematical Word Problem Generation from Commonsense Knowledge Graph and EquationsCode1
Evaluating Factuality in Generation with Dependency-level EntailmentCode1
Discern: Discourse-Aware Entailment Reasoning Network for Conversational Machine ReadingCode1
Inquisitive Question Generation for High Level Text ComprehensionCode1
Sequence-to-Sequence Learning for Indonesian Automatic Question GeneratorCode1
Can questions summarize a corpus? Using question generation for characterizing COVID-19 researchCode1
Text Generation by Learning from DemonstrationsCode1
A Dataset and Baselines for Visual Question Answering on ArtCode1
Investigating Pretrained Language Models for Graph-to-Text GenerationCode1
Visual Question Generation from Radiology ImagesCode1
ClarQ: A large-scale and diverse dataset for Clarification Question GenerationCode1
Show:102550
← PrevPage 4 of 27Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ERNIE-GENLARGE (beam size=5)BLEU-425.41Unverified
2BART (TextBox 2.0)BLEU-425.08Unverified
3ProphetNet + ASGenBLEU-424.44Unverified
4UniLMv2BLEU-424.43Unverified
5ProphetNet + syn. mask + localnessBLEU-424.37Unverified
6ProphetNetBLEU-423.91Unverified
7UniLM + ASGenBLEU-423.7Unverified
8UniLMBLEU-422.78Unverified
9BERTSQGBLEU-422.17Unverified
10Selector & NQG++BLEU-415.87Unverified
#ModelMetricClaimedVerifiedStatus
1MDNBLEU-165.1Unverified
2coco-Caption [[Karpathy and Li2014]]BLEU-162.5Unverified
3Max(Yang,2015)BLEU-159.4Unverified
4Sample(Yang,2015)BLEU-138.8Unverified
#ModelMetricClaimedVerifiedStatus
1FactJointGTMETEOR36.21Unverified
2JointGTMETEOR36.08Unverified
3FactT5BMETEOR35.72Unverified
4T5BMETEOR35.64Unverified
#ModelMetricClaimedVerifiedStatus
1FactT5BBLEU46.1Unverified
2JointGTBLEU45.95Unverified
3T5BBLEU44.51Unverified
4FactJointGTBLEU43.61Unverified
#ModelMetricClaimedVerifiedStatus
1JointGTMETEOR37.69Unverified
2FactJointGTMETEOR37.55Unverified
3FactT5BMETEOR37.39Unverified
4T5BMETEOR37.35Unverified
#ModelMetricClaimedVerifiedStatus
1BART fine-tuned on FairytaleQAROUGE-L0.53Unverified
2BART fine-tuned on NarrativeQA and FairytaleQAROUGE-L0.52Unverified
3BART fine-tuned on NarrativeQAROUGE-L0.44Unverified
#ModelMetricClaimedVerifiedStatus
1UniPollROUGE-149.6Unverified
2T5ROUGE-144.46Unverified
3Dual DecROUGE-138.24Unverified
#ModelMetricClaimedVerifiedStatus
1Info-HCVAEQAE37.18Unverified
2HCVAEQAE31.45Unverified
#ModelMetricClaimedVerifiedStatus
1Info-HCVAEQAE71.18Unverified
2HCVAEQAE69.46Unverified
#ModelMetricClaimedVerifiedStatus
1Info-HCVAEQAE35.45Unverified
2HCVAEQAE30.2Unverified
#ModelMetricClaimedVerifiedStatus
1MDNBLEU-136Unverified