Filters








11,911 Hits in 5.5 sec

Contextualized Sparse Representations for Real-Time Open-Domain Question Answering [article]

Jinhyuk Lee, Minjoon Seo, Hannaneh Hajishirzi, Jaewoo Kang
2020 arXiv   pre-print
Open-domain question answering can be formulated as a phrase retrieval problem, in which we can expect huge scalability and speed benefit but often suffer from low accuracy due to the limitation of existing  ...  In this paper, we aim to improve the quality of each phrase embedding by augmenting it with a contextualized sparse representation (Sparc).  ...  We thank the members of Korea University, University of Washington, NAVER Clova AI, and the anonymous reviewers for their insightful comments.  ... 
arXiv:1911.02896v2 fatcat:o4mojj5tjnhtjffqacpdpa72ey

Contextualized Sparse Representations for Real-Time Open-Domain Question Answering

Jinhyuk Lee, Minjoon Seo, Hannaneh Hajishirzi, Jaewoo Kang
2020 Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics   unpublished
Open-domain question answering can be formulated as a phrase retrieval problem, in which we can expect huge scalability and speed benefit but often suffer from low accuracy due to the limitation of existing  ...  In this paper, we aim to improve the quality of each phrase embedding by augmenting it with a contextualized sparse representation (SPARC).  ...  We thank the members of Korea University, University of Washington, NAVER Clova AI, and the anonymous reviewers for their insightful comments.  ... 
doi:10.18653/v1/2020.acl-main.85 fatcat:2xbfiyjiuzg2tkawpllnwrliha

Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Index

Minjoon Seo, Jinhyuk Lee, Tom Kwiatkowski, Ankur Parikh, Ali Farhadi, Hannaneh Hajishirzi
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
Existing open-domain question answering (QA) models are not suitable for real-time usage because they need to process several long documents on-demand for every input query, which is computationally prohibitive  ...  In this paper, we introduce query-agnostic indexable representations of document phrases that can drastically speed up open-domain QA.  ...  We thank the members of UW NLP, Google AI, and the anonymous reviewers for their insightful comments.  ... 
doi:10.18653/v1/p19-1436 dblp:conf/acl/SeoLKPFH19 fatcat:khayf34qmzhfjebghkopaurpmi

SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval [article]

Tiancheng Zhao, Xiaopeng Lu, Kyusong Lee
2020 arXiv   pre-print
We introduce SPARTA, a novel neural retrieval method that shows great promise in performance, generalization, and interpretability for open-domain question answering.  ...  We validated our approaches on 4 open-domain question answering (OpenQA) tasks and 11 retrieval question answering (ReQA) tasks.  ...  Conclusion In short, we propose SPARTA, a novel ranking method, that learns sparse representation for better open-domain QA.  ... 
arXiv:2009.13013v1 fatcat:qxsacmzhvrd3vjmsbf2tfu237a

NeuralQA: A Usable Library for Question Answering (Contextual Query Expansion + BERT) on Large Datasets [article]

Victor Dibia
2020 arXiv   pre-print
Existing tools for Question Answering (QA) have challenges that limit their use in practice.  ...  Code and documentation for NeuralQA is available as open source on Github (https://github.com/victordibia/neuralqaGithub).  ...  Acknowledgments The author thanks Melanie Beck, Andrew Reed, Chris Wallace, Grant Custer, Danielle Thorpe and other members of the Cloudera Fast Forward team for their valuable feedback.  ... 
arXiv:2007.15211v2 fatcat:q5wnqc6awzdw7nsxoyl4ehlb4u

Semantic Models for the First-stage Retrieval: A Comprehensive Review [article]

Yinqiong Cai, Yixing Fan, Jiafeng Guo, Fei Sun, Ruqing Zhang, Xueqi Cheng
2021 arXiv   pre-print
We believe it is the right time to survey current status, learn from existing methods, and gain some insights for future development.  ...  Moreover, we identify some open challenges and envision some future directions, with the hope of inspiring more researches on these important yet less investigated topics.  ...  approaches for open-domain question answering.  ... 
arXiv:2103.04831v3 fatcat:6qa7hvc3jve3pcmo2mo4qsiefq

Dynamic Adaptive Network Intelligence [article]

Richard Searle, Megan Bingham-Walker
2015 arXiv   pre-print
We report state-of-the-art results for DANI over question answering tasks in the bAbI dataset that have proved difficult for contemporary approaches to learning representation (Weston et al., 2015).  ...  Accurate representational learning of both the explicit and implicit relationships within data is critical to the ability of machines to perform more complex and abstract reasoning tasks.  ...  Although we report the application of DANI as an independent framework for learning representation, we recognize that our system could be employed to condition the input and intermediate layers of neural  ... 
arXiv:1511.06379v1 fatcat:2obpgubyjnfqhl7hwpw2tfh6qy

Adaptable Closed-Domain Question Answering Using Contextualized CNN-Attention Models and Question Expansion

Mahsa Abazari Kia, Aygul Garifullina, Mathias Kern, Jon Chamberlain, Shoaib Jameel
2022 IEEE Access  
In closed-domain Question Answering (QA), the goal is to retrieve answers to questions within a specific domain.  ...  Moreover, we include candidate answer identification and question expansion techniques for context reduction and rewriting ambiguous questions.  ...  [16] introduced Dense-Sparse Phrase Index (DENSPI), an indexable query-agnostic phrase representation model for real-time open-domain QA on SQuAD.  ... 
doi:10.1109/access.2022.3170466 fatcat:64rbm4tiqfb3hi4law253ioina

Adaptive Information Seeking for Open-Domain Question Answering [article]

Yunchang Zhu, Liang Pang, Yanyan Lan, Huawei Shen, Xueqi Cheng
2021 arXiv   pre-print
Information seeking is an essential step for open-domain question answering to efficiently gather evidence from a large corpus.  ...  In this paper, we propose a novel adaptive information-seeking strategy for open-domain question answering, namely AISO.  ...  Introduction Open-domain question answering (QA) (Voorhees et al., 1999) is a task of answering questions using a large collection of texts (e.g., Wikipedia).  ... 
arXiv:2109.06747v1 fatcat:n52qdsuzqjf3lpeexuosfuilh4

Dense-to-Question and Sparse-to-Answer: Hybrid Retriever System for Industrial Frequently Asked Questions

Jaehyung Seo, Taemin Lee, Hyeonseok Moon, Chanjun Park, Sugyeong Eo, Imatitikua D. Aiyanyo, Kinam Park, Aram So, Sungmin Ahn, Jeongbae Park
2022 Mathematics  
be continuously input for the FAQ system's maintenance.  ...  The term "Frequently asked questions" (FAQ) refers to a query that is asked repeatedly and produces a manually constructed response.  ...  Acknowledgments: Many thanks to KU NMT Group for taking the time to proofread this article. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/math10081335 fatcat:i2jlowhdgrd53hgegetw6uaozu

Accelerating Real-Time Question Answering via Question Generation [article]

Yuwei Fang, Shuohang Wang, Zhe Gan, Siqi Sun, Jingjing Liu, Chenguang Zhu
2021 arXiv   pre-print
To reduce the computational cost and accelerate real-time question answering (RTQA) for practical usage, we propose to remove all the neural networks from online QA systems, and present Ocean-Q (an Ocean  ...  Although deep neural networks have achieved tremendous success for question answering (QA), they are still suffering from heavy computational and energy cost for real product deployment.  ...  (i) We set up the first benchmark under the setting of answering open-domain questions without neural networks in real time.  ... 
arXiv:2009.05167v2 fatcat:sqy6s5f2jvbyfpltataeayoozm

Retrieving and Reading: A Comprehensive Survey on Open-domain Question Answering [article]

Fengbin Zhu, Wenqiang Lei, Chao Wang, Jianming Zheng, Soujanya Poria, Tat-Seng Chua
2021 arXiv   pre-print
Open-domain Question Answering (OpenQA) is an important task in Natural Language Processing (NLP), which aims to answer a question in the form of natural language based on large-scale unstructured documents  ...  We hope our work would enable researchers to be informed of the recent advancement and also the open challenges in OpenQA research, so as to stimulate further progress in this field.  ...  Specifically, textual QA is studied under two task settings based on the availability of contextual information, i.e. Machine Reading Comprehension (MRC) and Open-domain QA (OpenQA).  ... 
arXiv:2101.00774v3 fatcat:6evkg5cikjdp5fsi3ou3iqqkyq

Answering Questions on COVID-19 in Real-Time [article]

Jinhyuk Lee, Sean S. Yi, Minbyul Jeong, Mujeen Sung, Wonjin Yoon, Yonghwa Choi, Miyoung Ko, Jaewoo Kang
2020 arXiv   pre-print
answers to questions in real-time.  ...  We hope our system will be able to aid researchers in their search for knowledge and information not only for COVID-19, but for future pandemics as well.  ...  We thank the members of Korea University for the helpful comments and Kyle Lo (Allen Institute for Artificial Intelligence) for the insightful discussion.  ... 
arXiv:2006.15830v2 fatcat:36sbtup3rrgdtkqrqnw2i67qhi

Differentiable Reasoning over a Virtual Knowledge Base [article]

Bhuwan Dhingra, Manzil Zaheer, Vidhisha Balachandran, Graham Neubig, Ruslan Salakhutdinov, William W. Cohen
2020 arXiv   pre-print
We also describe a pretraining scheme for the contextual representation encoder by generating hard negative examples using existing knowledge bases.  ...  At each step the module uses a combination of sparse-matrix TFIDF indices and a maximum inner product search (MIPS) on a special index of contextual representations of the mentions.  ...  End-to-end open-domain question answering with bertserini. arXiv preprint arXiv:1902.01718, 2019. Zhilin Yang, Peng Qi, Saizheng Zhang, Yoshua Bengio, William W.  ... 
arXiv:2002.10640v1 fatcat:2fcnpovwjjfdvfl36nvhoj3u54

A Survey on Machine Reading Comprehension Systems [article]

Razieh Baradaran, Razieh Ghiasi, Hossein Amirkhani
2020 arXiv   pre-print
Its goal is to develop systems to answer the questions regarding a given context.  ...  Our investigations demonstrate that the focus of research has changed in recent years from answer extraction to answer generation, from single to multi-document reading comprehension, and from learning  ...  (A: Answer, P: passage, Q: Question) DATASET OPEN/CLOSE DOMAIN LANGUAGE QUESTION TYPE CONTEXT TYPE ANSWER TYPE #QUESTION #CONTEXT COLLECT DATA QUESTION CLASSIFICATION LINK ADDRESS  ... 
arXiv:2001.01582v2 fatcat:hb54svswpvgl5hpp5tesx2koca
« Previous Showing results 1 — 15 out of 11,911 results