scispace - formally typeset
Search or ask a question
Institution

Amazon.com

CompanySeattle, Washington, United States
About: Amazon.com is a company organization based out in Seattle, Washington, United States. It is known for research contribution in the topics: Service (business) & Service provider. The organization has 13363 authors who have published 17317 publications receiving 266589 citations.


Papers
More filters
Patent
27 Mar 2013
TL;DR: In this article, a user attempting to obtain information about an object can capture image information including a view of that object, and the image information can be used with a matching or identification process to provide information about that type of object to the user.
Abstract: A user attempting to obtain information about an object can capture image information including a view of that object, and the image information can be used with a matching or identification process to provide information about that type of object to the user. In order to narrow the search space to a specific category, and thus improve the accuracy of the results and the speed at which results can be obtained, the user can be guided to capture image information with an appropriate orientation. An outline or other graphical guide can be displayed over image information captured by a computing device, in order to guide the user in capturing the object from an appropriate direction and with an appropriate scale for the type of matching and/or information used for the matching. Such an approach enables three-dimensional objects to be analyzed using conventional two-dimensional identification algorithms, among other such processes.

68 citations

Proceedings ArticleDOI
01 Nov 2019
TL;DR: It is shown that data augmentation in feature space provides an effective way to improve intent classification performance in few-shot setting beyond traditional transfer learning approaches and upsampling in latent space is a competitive baseline for feature space augmentation.
Abstract: New conversation topics and functionalities are constantly being added to conversational AI agents like Amazon Alexa and Apple Siri. As data collection and annotation is not scalable and is often costly, only a handful of examples for the new functionalities are available, which results in poor generalization performance. We formulate it as a Few-Shot Integration (FSI) problem where a few examples are used to introduce a new intent. In this paper, we study six feature space data augmentation methods to improve classification performance in FSI setting in combination with both supervised and unsupervised representation learning methods such as BERT. Through realistic experiments on two public conversational datasets, SNIPS, and the Facebook Dialog corpus, we show that data augmentation in feature space provides an effective way to improve intent classification performance in few-shot setting beyond traditional transfer learning approaches. In particular, we show that (a) upsampling in latent space is a competitive baseline for feature space augmentation (b) adding the difference between two examples to a new example is a simple yet effective data augmentation method.

68 citations

Patent
15 Dec 2006
TL;DR: In this article, a method and system for determining and notifying users of undesirable network content are disclosed, where the adverse content event occurs dependent upon activity of a given user with respect to the given network information source.
Abstract: A method and system for determining and notifying users of undesirable network content are disclosed. According to one embodiment, a method may include detecting an adverse content event corresponding to a given network information source, where the adverse content event occurs dependent upon activity of a given user with respect to the given network information source. The method may also include reporting the adverse content event with respect to the given network information source, detecting a reference to the given network information source on behalf of a particular user, and in response to detecting the reference, retrieving an indication corresponding to the given network information source, where the indication is determined dependent upon adverse content events reported with respect to the given network information source. The method may further include notifying the particular user of possible undesirable content with respect to the given network information source dependent upon the indication.

68 citations

Journal ArticleDOI
03 Apr 2020
TL;DR: This work introduces MMM, a Multi-stage Multi-task learning framework for Multi-choice reading comprehension, and proposes a novel multi-step attention network (MAN) as the top-level classifier for this task.
Abstract: Machine Reading Comprehension (MRC) for question answering (QA), which aims to answer a question given the relevant context passages, is an important way to test the ability of intelligence systems to understand human language. Multiple-Choice QA (MCQA) is one of the most difficult tasks in MRC because it often requires more advanced reading comprehension skills such as logical reasoning, summarization, and arithmetic operations, compared to the extractive counterpart where answers are usually spans of text within given passages. Moreover, most existing MCQA datasets are small in size, making the task even harder. We introduce MMM, a Multi-stage Multi-task learning framework for Multi-choice reading comprehension. Our method involves two sequential stages: coarse-tuning stage using out-of-domain datasets and multi-task learning stage using a larger in-domain dataset to help model generalize better with limited data. Furthermore, we propose a novel multi-step attention network (MAN) as the top-level classifier for this task. We demonstrate MMM significantly advances the state-of-the-art on four representative MCQA datasets.

67 citations

Patent
Andrew Hayden1
26 Jan 2012
TL;DR: In this paper, a remote browsing and searching process is directed to the management of a remote browse session at a network computing provider, which provides search results corresponding to historical content representations associated with content previously requested by the client computing device, or to data about changes in the content.
Abstract: A remote browsing and searching process is directed to the management of a remote browse session at a network computing provider. Responsive to a search request, the network computing provider provides search results corresponding to historical content representations associated with content previously requested by the client computing device, search results corresponding to content representations associated with current content, or to data about changes in the content. The network computing provider may determine the search results based on a navigation path associated with a previous request for content, navigation paths of other client computing devices, relationships or differences between various versions of content, or based on any number of other factors. Interactive displays may be provided to client computing devices, allowing a user to refine the search results, zoom and manipulate content representations, and view relationships, similarities, and differences in content representations.

67 citations


Authors

Showing all 13498 results

NameH-indexPapersCitations
Jiawei Han1681233143427
Bernhard Schölkopf1481092149492
Christos Faloutsos12778977746
Alexander J. Smola122434110222
Rama Chellappa120103162865
William F. Laurance11847056464
Andrew McCallum11347278240
Michael J. Black11242951810
David Heckerman10948362668
Larry S. Davis10769349714
Chris M. Wood10279543076
Pietro Perona10241494870
Guido W. Imbens9735264430
W. Bruce Croft9742639918
Chunhua Shen9368137468
Network Information
Related Institutions (5)
Microsoft
86.9K papers, 4.1M citations

89% related

Google
39.8K papers, 2.1M citations

88% related

Carnegie Mellon University
104.3K papers, 5.9M citations

87% related

ETH Zurich
122.4K papers, 5.1M citations

82% related

University of Maryland, College Park
155.9K papers, 7.2M citations

82% related

Performance
Metrics
No. of papers from the Institution in previous years
YearPapers
20234
2022168
20212,015
20202,596
20192,002
20181,189