Filters








19,802 Hits in 8.6 sec

Character Matters: Video Story Understanding with Character-Aware Relations [article]

Shijie Geng, Ji Zhang, Zuohui Fu, Peng Gao, Hang Zhang, Gerard de Melo
2020 arXiv   pre-print
Video Story Question Answering (VSQA) offers an effective way to benchmark higher-level comprehension abilities of a model.  ...  We train and test our model on the six diverse TV shows in the TVQA dataset, which is by far the largest and only publicly available dataset for VSQA.  ...  of each modality for answering questions.  ... 
arXiv:2005.08646v1 fatcat:xuhhr4t6d5a2xly4muu4hc3mlm

Sentence Encoding with Tree-constrained Relation Networks [article]

Lei Yu, Cyprien de Masson d'Autume, Chris Dyer, Phil Blunsom, Lingpeng Kong, Wang Ling
2018 arXiv   pre-print
We instantiate this relational view of semantics in a series of neural models based on variants of relation networks (RNs) which represent a set of objects (for us, words forming a sentence) in terms of  ...  We propose two extensions to the basic RN model for natural language.  ...  Results on Quora duplicate questions detection. BiMPM and pt-DECATT (*) incorporate complex mechanism for mapping sentences; and the remainder of the models are sentence-encoding models.  ... 
arXiv:1811.10475v1 fatcat:ym3kkkrrvbdshgt6jjlrfrj3za

Exploring Answer Stance Detection with Recurrent Conditional Attention

Jianhua Yuan, Yanyan Zhao, Jingfang Xu, Bing Qin
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
RCA iteratively guides the distillation of question semantic with answer information and collects stance-oriented text relating to question, further revealing mutual relationship among stance, answer and  ...  Detecting stance from certain types of question-answer pairs is an interesting problem which has not been carefully explored.  ...  Acknowledgments We would like to thank all the anonymous reviewers for their insightful comments. Meanwhile, we thank Dr. Xiaocheng Feng, Prof. Wanxiang Che and Dr.  ... 
doi:10.1609/aaai.v33i01.33017426 fatcat:iljmnndpczhatemljsrr3kujya

Co-Stack Residual Affinity Networks with Multi-level Attention Refinement for Matching Text Sequences [article]

Yi Tay, Luu Anh Tuan, Siu Cheung Hui
2018 arXiv   pre-print
This task enables many potential applications such as question answering and paraphrase identification.  ...  Secondly, it leverages a multi-level attention refinement component between stacked recurrent layers.  ...  TrecQA (Wang et al., 2007) is a well-studied dataset for answer sentence selection task (or question-answer matching). The goal is to rank answers given a question.  ... 
arXiv:1810.02938v1 fatcat:setjh4bo45getdaj6ilaz7edvi

Opinion-aware Answer Generation for Review-driven Question Answering in E-Commerce [article]

Yang Deng, Wenxuan Zhang, Wai Lam
2020 arXiv   pre-print
Then a multi-view pointer-generator network is employed to generate opinion-aware answers for a given product-related question.  ...  the question and reviews to capture important information for answer generation, (ii) aggregating diverse opinion information to uncover the common opinion towards the given question.  ...  Finally, a multi-view pointer-generator network is exploited to combine the important information from both the question and reviews.  ... 
arXiv:2008.11972v2 fatcat:uncuhn3qrneo7jbdbb5by2sreq

Gaussian process decentralized data fusion meets transfer learning in large-scale distributed cooperative perception

Ruofei Ouyang, Bryan Kian Hsiang Low
2019 Autonomous Robots  
Yijun Wang*, Yingce Xia, Li Zhao, Jiang Bian, Tao Qin, Tie-Yan Liu, Guiquan Liu Dual-reference Face Retrieval BingZhang Hu, Feng Zheng, Ling Shao* Duplicate Question Identification by Integrating FrameNet  ...  Qin, Chandini Shetty HCVRD: a benchmark for large-scale Human-Centered Visual Relationship Detection Bohan Zhuang, Qi Wu*, Ian Reid, Chunhua Shen, Anton van den Hengel Hierarchical Attention Flow for  ... 
doi:10.1007/s10514-018-09826-z fatcat:67yqhwmgozccxni56rxmuapjgm

Visual Question Answering Using Semantic Information from Image Descriptions [article]

Tasmia Tasrin, Md Sultan Al Nahian, Brent Harrison
2021 arXiv   pre-print
the regions of an image to produce open-ended answers for questions asked in a visual question answering (VQA) task.  ...  In this work, we propose a deep neural architecture that uses an attention mechanism which utilizes region based image features, the natural language question asked, and semantic knowledge extracted from  ...  Our resulting network, which we call the Visual Question Answering-Contextual Information network (VQA-CoIn), improves upon past work by extending it to incorporate semantic information extracted from  ... 
arXiv:2004.10966v2 fatcat:ycgvdsknffe3xcx4unbmvoluly

Visual Question Answering Using Semantic Information from Image Descriptions

Tasmia Tasmia, Md Sultan Al Nahian, Brent Harrison
2021 Proceedings of the ... International Florida Artificial Intelligence Research Society Conference  
the regions of an image to produce open-ended answers for questions asked in a visual question answering (VQA) task.  ...  In this work, we propose a deep neural architecture that uses an attention mechanism which utilizes region based image features, the natural language question asked, and semantic knowledge extracted from  ...  Our resulting network, which we call the Visual Question Answering-Contextual Information network (VQA-CoIn), improves upon past work by extending it to incorporate semantic information extracted from  ... 
doi:10.32473/flairs.v34i1.128460 fatcat:v5exd2g4hbecbdug3k2ilxltbu

Towards information-rich, logical text generation with knowledge-enhanced neural models [article]

Hao Wang, Bin Guo, Wei Wu, Zhiwen Yu
2020 arXiv   pre-print
The relation detection module calculates similarity scores of each question and its relation candidates to select the triple with highest score to answer the question.  ...  The GCN is used to capture reasoning chains by propagating local contextual information along edges to perform multi-step reasoning for generating answers. Lv et al.  ... 
arXiv:2003.00814v1 fatcat:5fllyakwqzf4vnmar3a6zjoewe

Deep learning approaches to pattern extraction and recognition in paintings and drawings: an overview

Giovanna Castellano, Gennaro Vessio
2021 Neural computing & applications (Print)  
Recent advances in deep learning and computer vision, coupled with the growing availability of large digitized visual art collections, have opened new opportunities for computer science researchers to  ...  Visual question answering Recently, Garcia at al.  ...  [54] Visual question answering Ufer et al.  ... 
doi:10.1007/s00521-021-05893-z fatcat:elqzw3hzbzgodotie6ndih537u

Enhancements to the Sequence-to-Sequence Based Natural Answer Generation Models

Kulothunkan Palasundram, Nurfadhlina Mohd Sharef, Khairul Azhar Kasmiran, Azreen Azman
2020 IEEE Access  
For more information, see https://creativecommons.org/licenses/by/4.0/ VOLUME 8, 2020 K. Palasundram et al.: Enhancements to the Seq2Seq-Based NAG Models VOLUME 8, 2020  ...  The Seq2Seq model shows a weakness whereby the model tends to generate answers that are generic, meaningless and inconsistent with the questions.  ...  ACKNOWLEDGMENT The authors would like to thank AOARD for the support and express our appreciation to all who have contributed to this research.  ... 
doi:10.1109/access.2020.2978551 fatcat:qynhwyjwqjdubpuqq2ym7bdvta

Framework for Deep Learning-Based Language Models using Multi-task Learning in Natural Language Understanding: A Systematic Literature Review and Future Directions

Rahul Manohar Samant, Mrinal Bachute, Shilpa Gite, Ketan Kotecha
2022 IEEE Access  
Even though MTL (Multi-task Learning) was introduced before Deep Learning, it has gained significant attention in the past years.  ...  However, they face the challenge of designing the generalpurpose framework for the language model, which will improve the performance of multi-task NLU and the generalized representation of knowledge.  ...  A novel multi-tasks question answering network (MQAN) is proposed, which learns decaNLP tasks without fine-tuning any network [74] .  ... 
doi:10.1109/access.2022.3149798 fatcat:k3kdt4eryzdfpk5k6w62jtlskm

A Survey on Dialogue Systems: Recent Advances and New Frontiers [article]

Hongshen Chen, Xiaorui Liu, Dawei Yin, Jiliang Tang
2018 arXiv   pre-print
Dialogue systems have attracted more and more attention.  ...  For dialogue systems, deep learning can leverage a massive amount of data to learn meaningful feature representations and response generation strategies, while requiring a minimum amount of hand-crafting  ...  [123] adopted an encoder-decoder LSTM-based structure to incorporate the question information, semantic slot values, and dialogue act type to generate correct answers.  ... 
arXiv:1711.01731v3 fatcat:6wuovcynqbhlzmuorchn4mn6ma

Learning Future Object Prediction with a Spatiotemporal Detection Transformer [article]

Adam Tonderski, Joakim Johnander, Christoffer Petersson, Kalle Åström
2022 arXiv   pre-print
Second, we feed ego-motion information to the model via cross-attention. We show that both of these cues substantially improve future object prediction performance.  ...  Our final approach learns to capture the dynamics and make predictions on par with an oracle for 100 ms prediction horizons, and outperform baselines for longer prediction horizons.  ...  The computations were enabled by resources provided by the Swedish National Infrastructure for Computing (SNIC), partially funded by the Swedish Research Council through grant agreement no. 2018-05973.  ... 
arXiv:2204.10321v1 fatcat:35bkglo3xvf6lkdochxr3idkuq

Attention, please! A survey of Neural Attention Models in Deep Learning [article]

Alana de Santana Correia, Esther Luna Colombini
2021 arXiv   pre-print
For the last six years, this property has been widely explored in deep neural networks.  ...  Given our limited ability to process competing sources, attention mechanisms select, modulate, and focus on the information most relevant to behavior.  ...  Attention Network [516] for question answering tasks. a) The show, attend and tell framework.  ... 
arXiv:2103.16775v1 fatcat:lwkw42lrircorkymykpgdmlbwq
« Previous Showing results 1 — 15 out of 19,802 results