Share on Facebook Tweet on Twitter Share on LinkedIn Share by email
Our research
Content type
+
Downloads (456)
+
Events (484)
 
Groups (150)
+
News (2840)
 
People (723)
 
Projects (1157)
+
Publications (12944)
+
Videos (6109)
Labs
Research areas
Algorithms and theory47205 (375)
Communication and collaboration47188 (243)
Computational linguistics47189 (267)
Computational sciences47190 (245)
Computer systems and networking47191 (843)
Computer vision208594 (948)
Data mining and data management208595 (154)
Economics and computation47192 (127)
Education47193 (91)
Gaming47194 (85)
Graphics and multimedia47195 (255)
Hardware and devices47196 (235)
Health and well-being47197 (113)
Human-computer interaction47198 (1007)
Machine learning and intelligence47200 (1022)
Mobile computing208596 (88)
Quantum computing208597 (41)
Search, information retrieval, and knowledge management47199 (749)
Security and privacy47202 (364)
Social media208598 (91)
Social sciences47203 (313)
Software development, programming principles, tools, and languages47204 (679)
Speech recognition, synthesis, and dialog systems208599 (178)
Technology for emerging markets208600 (64)
1–25 of 267
Sort
Show 25 | 50 | 100
1234567Next 
Kenneth Tran, Xiaodong He, Lei Zhang, Jian Sun, Cornelia Carapcea, Chris Thrasher, Chris Buehler, and Chris Sienkiewicz

We present an image caption system that addresses new challenges of automatically describing images in the wild. The challenges include high quality caption quality with respect to human judgments, out-of-domain data handling, and low latency required in many applications. Built on top of a state-of-the-art framework, we developed a deep vision model that detects a broad range of visual concepts, an entity recognition model that identifies celebrities and landmarks, and a confidence model for the...

Publication details
Date: 1 June 2016
Type: Article
Moontae Lee, Xiaodong He, Wen-tau Yih, Jianfeng Gao, Li Deng, and Paul Smolensky

Question answering tasks have shown remarkable progress with distributed vector representation. In this paper, we investigate the recently proposed Facebook bAbI tasks which consist of twenty different categories of questions that require complex reasoning. Because the previous work on bAbI are all end-to-end models, errors could come from either an imperfect understanding of semantics or in certain steps of the reasoning. For clearer analysis, we propose two vector space models inspired by Tensor...

Publication details
Date: 2 May 2016
Type: Inproceeding
Huan Sun, Hao Ma, Xiaodong He, Wen-tau Yih, Yu Su, and Xifeng Yan

Tables are pervasive on the Web. Informative web tables range across a large variety of topics, which can naturally serve as a significant resource to satisfy user information needs. Driven by such observations, in this paper, we investigate an important yet largely under-addressed problem: Given millions of tables, how to precisely retrieve table cells to answer a user question. This work proposes a novel table cell search framework to attack this problem. We first formulate the concept of a...

Publication details
Date: 11 April 2016
Type: Inproceeding
Publisher: ACM – Association for Computing Machinery
Publication details
Date: 1 April 2016
Type: Technical report
Publisher: MS tech report
Number: MSR-TR-2016-16
Ting-Hao Huang, Francis Ferraro, Nasrin Mostafazadeh, Ishan Misra, Aishwarya Agrawal, Jacob Devlin, Ross Girshick, Xiaodong He, Pushmeet Kohli, Dhruv Batra, C. Lawrence Zitnick, Devi Parikh, Lucy Vanderwende, Michel Galley, and Margaret Mitchell

We introduce the first dataset for sequential vision-to-language, and explore how this data may be used for the task of visual storytelling. The first release of this dataset, SIND1 v.1, includes 81,743 unique photos in 20,211 sequences, aligned to both descriptive (caption) and story language. We establish several strong baselines for the storytelling task, and motivate an automatic metric to benchmark progress. Modelling concrete description as well as figurative and social language, as...

Publication details
Date: 1 April 2016
Type: Inproceeding
Publisher: ACL – Association for Computational Linguistics
Jiwei Li, Michel Galley, Chris Brockett, Jianfeng Gao, and Bill Dolan

Sequence-to-sequence neural network models for generation of conversational responses tend to generate safe, commonplace responses (e.g., I don't know) regardless of the input. We suggest that the traditional objective function, i.e., the likelihood of output (response) given input (message) is unsuited to response generation tasks. Instead we propose using Maximum Mutual Information (MMI) as the objective function in neural models. Experimental results demonstrate that the proposed MMI models...

Publication details
Date: 1 March 2016
Type: Proceedings
Shi Feng, Shujie Liu, Mu Li, and Ming Zhou

Neural machine translation has shown very promising results lately.Most NMT models follow the encoder-decoder framework. To make encoder-decoder models more flexible, attention mechanism was introduced to machine translation and also other tasks like speech recognition and image captioning. We observe that the quality of translation by attention-based encoder-decoder can be significantly damaged when the alignment is incorrect. We attribute these problems to the lack of distortion and fertility models....

Publication details
Date: 1 February 2016
Type: Technical report
Number: MSR-TR-2016-11
William D. Lewis, Christian Federmann, and Ying Xin

Cross Entropy Difference (CED) has proven to be a very effective method for selecting domain-specific data from large corpora of out-of-domain or general domain content. It is used in a number of different scenarios, and is particularly popular in bake-off competitions in which participants have a limited set of resources to draw from, and need to sub-sample the data in such a way as to ensure better results on domain-specific test sets. The underlying algorithm is handy since one can provide a set of...

Publication details
Date: 4 December 2015
Type: Inproceeding
Yun-Nung Chen, Dilek Hakkani-Tur, and Xiaodong He

The recent success of voice interaction with smart devices (humanmachine genre) and improvements in speech recognition for conversational speech show the possibility of conversation-related applications. This paper investigates the task of actionable item detection in meetings (human-human genre), where the intelligent assistant dynamically provides the participants access to information (e.g. scheduling a meeting, taking notes) without interrupting the meetings. A convolutional deep structured semantic...

Publication details
Date: 1 December 2015
Type: Inproceeding
Publisher: IEEE – Institute of Electrical and Electronics Engineers
Sauleh Eetemadi, William Lewis, Kristina Toutanova, and Hayder Radha

Statistical machine translation has seen significant improvements in quality over the past several years. The single biggest factor in this improvement has been the accumulation of ever larger stores of data. We now find ourselves, however, the victims of our own success, in that it has become increasingly difficult to train on such large sets of data, due to limitations in memory, processing power, and ultimately, speed (i.e. data-to-models takes an inordinate amount of time). Moreover, the training...

Publication details
Date: 1 December 2015
Type: Article
Publisher: Springer
Royal Sequiera, Monojit Choudhury, and Kalika Bali

We discuss Part-of-Speech(POS) tagging of Hindi-English Code-Mixed(CM) text from social media content. We propose extensions to the existing approaches, we also present a new feature set which addresses the transliteration problem inherent in social media. We achieve an 84% accuracy with the new feature set. We show that the context and joint modeling of language detection and POS tag layers do not help in POS tagging.

Publication details
Date: 1 December 2015
Type: Inproceeding
Publisher: NLPAI
Royal Sequiera, Monojit Choudhury, Parth Gupta, Paolo Rosso, Shubham Kumar, Somnath Banerjee, Sudip Kumar Naskar, Sivaji Bandyopadhyay, Gokul Chittaranjan, Amitava Das, and Kunal Chakma

The Transliterated Search track has been organized for the third year in FIRE-2015. The track had three subtasks. Subtask I was on language labeling of words in code-mixed text fragments; it was conducted for 8 Indian languages: Bangla, Gujarati, Hindi, Kannada, Malayalam, Marathi, Tamil, Telugu, mixed with English. Subtask II was on ad-hoc retrieval of Hindi film lyrics, movie reviews and astrology documents, where both the queries and documents were either in Hindi written in Devanagari or in Roman...

Publication details
Date: 1 December 2015
Type: Inproceeding
Publisher: FIRE
William D. Lewis

In 1966, Star Trek introduced us to the notion of the Universal Translator. Such a device allowed Captain Kirk and his crew to communicate with alien species, such as the Gorn, who did not speak their language, or even converse with species who did not speak at all (e.g., the Companion from the episode Metamorphosis). In 1979, Douglas Adams introduced us to the “Babelfish” in the Hitchhiker's Guide to the Galaxy which, when inserted into the ear, allowed the main character to do...

Publication details
Date: 27 November 2015
Type: Inproceeding
Qinglin Li, Shujie Liu, Rui Lin, Mu Li, and Ming Zhou

Nowadays knowledge base (KB) has been viewed as one of the im-
portant infrastructures for many web search applications and NLP tasks. How-
ever, in practice the availability of KB data varies from language to language,
which greatly limits potential usage of knowledge base. In this paper, we pro-
pose a novel method to construct or enrich a knowledge base by entity translation
with help of another KB but compiled in a different language. In our work, we
concentrate on two...

Publication details
Date: 1 October 2015
Type: Proceedings
Publisher: NLPCC
Yi Yang, Wen-tau Yih, and Christopher Meek

We describe the WikiQA dataset, a new publicly available set of question and sentence pairs, collected and annotated for research on open-domain question answering. Most previous work on answer sentence selection focuses on a dataset created using the TREC-QA data, which includes editor-generated questions and candidate answer sentences selected by matching content words in the question. WikiQA is constructed using a more natural process and is more than an order of magnitude larger than the previous...

Publication details
Date: 21 September 2015
Type: Inproceeding
Publisher: ACL – Association for Computational Linguistics
Kristina Toutanova, Danqi Chen, Patrick Pantel, Hoifung Poon, Pallavi Choudhury, and Michael Gamon

Models that learn to represent textual and knowledge base relations in the same continuous latent space are able to perform joint inferences among the two kinds of relations and obtain high accuracy on knowledge base completion (Riedel et al. 2013). In this paper we propose a model that captures the compositional structure of textual relations, and jointly optimizes entity, knowledge base, and textual relation representations. The proposed model significantly improves performance over a model that...

Publication details
Date: 17 September 2015
Type: Inproceeding
Publisher: ACL – Association for Computational Linguistics
Paul A. Crook, Jean-Philippe Robichaud, and Ruhi Sarikaya

Hypothesis ranking (HR) is an approach for improving the accuracy of both domain detection and tracking in multi-domain, multi-turn dialogue systems. This paper presents the results of applying a universal HR model to multiple dialogue systems, each of which are using a different language. It demonstrates that as the set of input features used by HR models are largely language independent a single, universal HR model can be used in place of language specific HR models with only a small loss in accuracy...

Publication details
Date: 1 September 2015
Type: Inproceeding
Publisher: ISCA - International Speech Communication Association
Daniel Preotiuc-Pietro, Svitlana Volkova, Vasileios Lampos, Yoram Bachrach, and Nikolaos Aletras

Automatically inferring user demographics from social media posts is useful for both social science research and a range of downstream applications in marketing and politics. We present the first extensive study where user behaviour on Twitter is used to build a predictive model of income. We apply non-linear methods for regression, i.e. Gaussian Processes, achieving strong correlation between predicted and actual user income. This allows us to shed light on the factors that characterise income on...

Publication details
Date: 1 September 2015
Type: Article
Publisher: PLOS – Public Library of Science
Rui Lin, Shujie Liu, Muyun Yang, Mu Li, Ming Zhou, and Sheng Li

This paper proposes a novel hierarchical recurrent neural network language model (HRNNLM) for document modeling. After establishing a RNN to capture the coherence between sentences in a document, HRNNLM integrates it as the sentence history information into the word level RNN to predict the word sequence with cross-sentence contextual information. A two-step training approach is designed, in which sentence-level and word-level language models are approximated for the convergence in a pipeline style....

Publication details
Date: 1 September 2015
Type: Inproceeding
Publisher: EMNLP
Nicholas Ruiz, Qin Gao, William Lewis, and Marcello Federico

In the spoken language translation pipeline, machine translation systems that are trained solely on written bitexts are often unable to recover from speech recognition errors due to the mismatch in training data. We propose a novel technique to simulate the errors generated by an ASR system, using the ASR system’s pronunciation dictionary and language model. Lexical entries in the pronunciation dictionary are converted into phoneme sequences using a text-to-speech (TTS) analyzer and stored in a...

Publication details
Date: 1 September 2015
Type: Inproceeding
Publisher: ISCA - International Speech Communication Association
Dilek Hakkani-Tur, Yun-Cheng Ju, Geoffrey Zweig, and Gokhan Tur

Spoken language understanding (SLU) in today’s conversational systems focuses on recognizing a set of domains, intents, and associated arguments, that are determined by application developers. User requests that are not covered by these are usually directed to search engines, and may remain unhandled. We propose a method that aims to find common user intents amongst these uncovered, out-of-domain utterances, with the goal of supporting future phases of dialog system design. Our approach relies on...

Publication details
Date: 1 September 2015
Type: Inproceeding
Publisher: Interspeech 2015 Conference
Young-Bum Kim, Karl Stratos, Ruhi Sarikaya, and Minwoo Jeong

In natural language understanding (NLU), a user utterance can be labeled differently depending on the domain or application (e.g., weather vs. calendar). Standard domain adaptation techniques are not directly applicable to take advantage of the existing annotations because they assume that the label set is invariant. We propose a solution based on label embeddings induced from canonical correlation analysis (CCA) that reduces the problem to a standard domain adaptation task and allows use of a number of...

Publication details
Date: 29 August 2015
Type: Proceedings
Publisher: ACL – Association for Computational Linguistics
Young-Bum Kim, Karl Stratos, and Ruhi Sarikaya

In this paper, we apply the concept of pre-training to hidden-unit conditional random
fields (HUCRFs) to enable learning on unlabeled data. We present a simple yet effective pre-training technique that learns to associate words with their clusters, which are obtained in an unsupervised manner. The learned parameters are then used to initialize the supervised learning process. We also propose a word clustering technique based on canonical correlation analysis (CCA) that is sensitive to multiple word...

Publication details
Date: 28 August 2015
Type: Proceedings
Publisher: ACL – Association for Computational Linguistics
Young-Bum Kim, Karl Stratos, Xiaohu Liu, and Ruhi Sarikaya

In this paper, we introduce the task of selecting compact lexicon from large, noisy gazetteers.
This scenario arises often in practice, in particular spoken language understanding (SLU).
We propose a simple and effective solution based on matrix decomposition techniques:
canonical correlation analysis (CCA) and rank-revealing QR (RRQR) factorization. CCA is first used to derive low-dimensional gazetteer embeddings from domain-specific search logs. Then RRQR is used to find a subset of...

Publication details
Date: 27 August 2015
Type: Proceedings
Publisher: ACL – Association for Computational Linguistics
Jian Tang, Meng Qu, and Qiaozhu Mei

Unsupervised text embedding methods, such as Skip-gram and Paragraph Vector, have been attracting increasing attention due to their simplicity, scalability, and effectiveness. However, comparing to sophisticated deep learning architectures such as convolutional neural networks, these methods usually yield inferior results when applied to particular machine learning tasks. One possible reason is that these text embedding methods learn the representation of text in a fully unsupervised way, without...

Publication details
Date: 1 August 2015
Type: Inproceeding
Publisher: ACM – Association for Computing Machinery
1–25 of 267
Sort
Show 25 | 50 | 100
1234567Next 
> Our research