31,540 Hits in 7.7 sec

Knowledge-enriched, Type-constrained and Grammar-guided Question Generation over Knowledge Bases [article]

Sheng Bi and Xiya Cheng and Yuan-Fang Li and Yongzhen Wang and Guilin Qi
2020 arXiv   pre-print
Question generation over knowledge bases (KBQG) aims at generating natural-language questions about a subgraph, i.e. a set of (connected) triples.  ...  In our model, the encoder is equipped with auxiliary information from the KB, and the decoder is constrained with word types during QG.  ...  For example, KBQG can improve factoid-based question answering (QA) systems by either dual training of QA and QG or by data augmentation for training corpora.  ... 
arXiv:2010.03157v3 fatcat:24iss66z45fkfgdan3nfuh5ol4

Co-attending Free-form Regions and Detections with Multi-modal Multiplicative Feature Embedding for Visual Question Answering [article]

Pan Lu, Hongsheng Li, Wei Zhang, Jianyong Wang, Xiaogang Wang
2017 arXiv   pre-print
question-related free-form image regions and detection boxes for more accurate question answering.  ...  Existing VQA methods mainly adopt the visual attention mechanism to associate the input question with corresponding image regions for effective question answering.  ...  The models in the first part of Table 1 are based on simple joint question-image feature embedding methods.  ... 
arXiv:1711.06794v2 fatcat:wdnfot2anzfsvabke4pykycxne

Co-Attending Free-Form Regions and Detections With Multi-Modal Multiplicative Feature Embedding for Visual Question Answering

Pan Lu, Hongsheng Li, Wei Zhang, Jianyong Wang, Xiaogang Wang
question-related free-form image regions and detection boxes for more accurate question answering.  ...  Existing VQA methods mainly adopt the visual attention mechanism to associate the input question with corresponding image regions for effective question answering.  ...  The models in the first part of Table 1 are based on simple joint question-image feature embedding methods.  ... 
doi:10.1609/aaai.v32i1.12240 fatcat:7bw2ltvyxzdq5ddktea72kj4by

Question Answering and Question Generation as Dual Tasks [article]

Duyu Tang, Nan Duan, Tao Qin, Zhao Yan, Ming Zhou
2017 arXiv   pre-print
We study the problem of joint question answering (QA) and question generation (QG) in this paper.  ...  On one side, the QA model judges whether the generated question of a QG model is relevant to the answer.  ...  We sincerely thank Wenpeng Yin for running the powerful ABCNN model on our setup.  ... 
arXiv:1706.02027v2 fatcat:4prdxbgdbbdkfagneqljub5qx4

DiaLex: A Benchmark for Evaluating Multidialectal Arabic Word Embeddings [article]

Muhammad Abdul-Mageed, Shady Elbassuoni, Jad Doughman, AbdelRahim Elmadany, El Moatez Billah Nagoudi, Yorgo Zoughby, Ahmad Shaher, Iskander Gaba, Ahmed Helal, Mohammed El-Razzaz
2021 arXiv   pre-print
Across these dialects, DiaLex provides a testbank for six syntactic and semantic relations, namely male to female, singular to dual, singular to plural, antonym, comparative, and genitive to past tense  ...  We describe DiaLex, a benchmark for intrinsic evaluation of dialectal Arabic word embedding. DiaLex covers five important Arabic dialects: Algerian, Egyptian, Lebanese, Syrian, and Tunisian.  ...  That is, assume the number of questions for a given dialect and a given relation is n, and assume that a given embeddings model M correctly answered m out of those n questions as explained above.  ... 
arXiv:2011.10970v2 fatcat:o23appv335dpzm7uixyy2ynouy

Uncovering the Temporal Context for Video Question Answering

Linchao Zhu, Zhongwen Xu, Yi Yang, Alexander G. Hauptmann
2017 International Journal of Computer Vision  
We explore approaches for finer understanding of video content using question form of "fill-in-the-blank", and managed to collect 109,895 video clips with duration over 1,000 hours from TACoS, MPII-MD,  ...  We present an encoder-decoder approach using Recurrent Neural Networks to learn temporal structures of videos and introduce a dual-channel ranking loss to answer multiple-choice questions.  ...  Video Question Answering and temporal structure reasoning. To the best of our knowledge, the only work on video-based question answering is Tu et al.  ... 
doi:10.1007/s11263-017-1033-7 fatcat:5or4ebm2inbc7faqhxzklsvnaq

Capturing and modeling the process of conceptual change

Stella Vosniadou
1994 Learning and Instruction  
Some kinds of conceptual change require the simple addition of new information to an existing conceptual structure. Others are accomplished only when existing beliefs and presuppositions are revised.  ...  Because generative questions cannot be answered through the simple repetition of unassimilated information, they have a greater potential for unraveling underlying conceptual structures.  ...  They then use this model to answer the question.  ... 
doi:10.1016/0959-4752(94)90018-3 fatcat:tgmmy6c7ivc75bxtlnktpaytwq

SG-Net: Syntax Guided Transformer for Language Representation [article]

Zhuosheng Zhang, Yuwei Wu, Junru Zhou, Sufeng Duan, Hai Zhao, Rui Wang
2021 arXiv   pre-print
In detail, for self-attention network (SAN) sponsored Transformer-based encoder, we introduce syntactic dependency of interest (SDOI) design into the SAN to form an SDOI-SAN with syntax-guided self-attention  ...  For language representation, the capacity of effectively modeling the linguistic knowledge from the detail-riddled and lengthy texts and getting rid of the noises is essential to improve its performance  ...  simple linear layers) in specific downstream NLP tasks (e.g., text classification, question answering, natural language inference).  ... 
arXiv:2012.13915v2 fatcat:2zyyd4s6ibcuvjal3k7t2e4v44

Semantic Structure based Query Graph Prediction for Question Answering over Knowledge Graph [article]

Mingchen Li, Shihao Ji
2022 arXiv   pre-print
Building query graphs from natural language questions is an important step in complex question answering over knowledge graph (Complex KGQA).  ...  By doing so, we can first filter out noisy candidate query graphs by the predicted semantic structures, and then rank the remaining candidates with a BERT-based ranking model.  ...  Users can get crisp answers by querying KGs with natural language questions, and this process is called Question Answering over Knowledge Graph (KGQA) .  ... 
arXiv:2204.10194v2 fatcat:h67zwsa4yjd4pmbxzhsm3m6tx4

Learning Word Vectors with Linear Constraints: A Matrix Factorization Approach

Wenye Li, Jiawei Zhang, Jianjun Zhou, Laizhong Cui
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
With the objective of better capturing the semantic and syntactic information inherent in words, we propose two new embedding models based on the singular value decomposition of lexical co-occurrences  ...  Different from previous work, our proposed models allow for injecting linear constraints when performing the decomposition, with which the desired semantic and syntactic information will be maintained  ...  For our additive model, 1%, 2%, 3% of questions (with answers) were randomly chosen and used as the prior knowledge during training.  ... 
doi:10.24963/ijcai.2018/582 dblp:conf/ijcai/LiZZC18 fatcat:jlzhuxfynfdmln64cvjekt3pgy

Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering

Yi Tay, Luu Anh Tuan, Siu Cheung Hui
2018 Proceedings of the Eleventh ACM International Conference on Web Search and Data Mining - WSDM '18  
We propose a simple but novel deep learning architecture for fast and efficient question-answer ranking and retrieval.  ...  The dominant neural architectures in question answer retrieval are based on recurrent or convolutional encoders configured with complex word matching layers.  ...  Figure 5 : 5 Histogram plots of embedding norms. Figure 6 : 6 Dual hierarchical matching of question and answer at word-level. .  ... 
doi:10.1145/3159652.3159664 dblp:conf/wsdm/TayTH18 fatcat:ebv4xtt4jzgdnmpbld4i535xpi

Enabling Multimodal Generation on CLIP via Vision-Language Knowledge Distillation [article]

Wenliang Dai, Lu Hou, Lifeng Shang, Xin Jiang, Qun Liu, Pascale Fung
2022 arXiv   pre-print
To tackle this problem, we propose to augment the dual-stream VLP model with a textual pre-trained language model (PLM) via vision-language knowledge distillation (VLKD), enabling the capability for multimodal  ...  Experimental results show that the resulting model has strong zero-shot performance on multimodal generation tasks, such as open-ended visual question answering and image captioning.  ...  Related Work Vision-language Pre-training Based on how the two modalities interact, recent VLP models mainly fall into two categories: singlestream and dual-stream models.  ... 
arXiv:2203.06386v2 fatcat:oi6r6xjmofeold7dnku2steab4

Code Generation as a Dual Task of Code Summarization [article]

Bolin Wei, Ge Li, Xin Xia, Zhiyi Fu, Zhi Jin
2019 arXiv   pre-print
Various neural network-based approaches are proposed to solve these two tasks separately.  ...  In this paper, we apply the relations between two tasks to improve the performance of both tasks.  ...  Acknowledgments We thank all reviewers for their constructive comments, Fang Liu for discussion on manuscript.  ... 
arXiv:1910.05923v1 fatcat:345xb56okba67gz3bitteq5ozi

Reference Knowledgeable Network for Machine Reading Comprehension [article]

Yilin Zhao, Zhuosheng Zhang, Hai Zhao
2022 arXiv   pre-print
Multi-choice Machine Reading Comprehension (MRC) as a challenge requires models to select the most appropriate answer from a set of candidates with a given passage and question.  ...  make up for the deficiency of the given passage.  ...  Feng [41] quoted multi-hop commonsense for simple questions without relevant contexts, while Lin [42] proposed a constrained text generation task for generative commonsense reasoning.  ... 
arXiv:2012.03709v3 fatcat:ljp3yhhkfbclfpegu6uoy3tqj4

Text-based Question Answering from Information Retrieval and Deep Neural Network Perspectives: A Survey [article]

Zahra Abbasiantaeb, Saeedeh Momtazi
2020 arXiv   pre-print
Text-based Question Answering (QA) is a challenging task which aims at finding short concrete answers for users' questions.  ...  This line of research has been widely studied with information retrieval techniques and has received increasing attention in recent years by considering deep neural network approaches.  ...  There are two major approaches for QA systems: text-based QA, and knowledge-based QA. Knowledge-based QAs rely on knowledge bases (KB) for finding the answer to the user's question.  ... 
arXiv:2002.06612v2 fatcat:wndvk257bncfphgtzxz7sr4o4e
« Previous Showing results 1 — 15 out of 31,540 results