Filters








872 Hits in 5.0 sec

Learning perceptually grounded word meanings from unaligned parallel data

Stefanie Tellex, Pratiksha Thaker, Joshua Joseph, Nicholas Roy
2013 Machine Learning  
In order for robots to effectively understand natural language commands, they must be able to acquire meaning representations that can be mapped to perceptual features in the external world.  ...  In this paper, we present an approach to grounded language acquisition which is capable of jointly learning a policy for following natural language commands such as "Pick up the tire pallet," as well as  ...  We would also like to thank Thomas Howard for his helpful comments on a draft of this paper, and Matthew R. Walter for his help in collecting the corpus.  ... 
doi:10.1007/s10994-013-5383-2 fatcat:zxapycqw7fa77f6pesia3oveky

Graph-Based Semi-Supervised Conditional Random Fields For Spoken Language Understanding Using Unaligned Data [article]

Mohammad Aliannejadi, Masoud Kiaeeha, Shahram Khadivi, Saeed Shiry Ghidary
2017 arXiv   pre-print
We experiment graph-based Semi-Supervised Learning (SSL) of Conditional Random Fields (CRF) for the application of Spoken Language Understanding (SLU) on unaligned data.  ...  Our results demonstrate that our proposed approach significantly improves the performance of the supervised model by utilizing the knowledge gained from the graph.  ...  Semi-supervised Spoken Language Understanding The input data is unaligned and represented as a semantic tree, which is described in (He and Young, 2005) .  ... 
arXiv:1701.08533v1 fatcat:4fpljknhxrbz5pv37nq442buuq

MEDT: Using Multimodal Encoding-Decoding Network as in Transformer for Multimodal Sentiment Analysis

Qingfu Qi, Liyuan Lin, Rui Zhang, Chengrong Xue
2022 IEEE Access  
131" Innovative Talent Team.The author sincerely thanks the above teams for their support to our research.  ...  It models natural language and non-natural language to gain emotional understanding.  ...  language data caused by non-natural language data.  ... 
doi:10.1109/access.2022.3157712 fatcat:rjsmtx2ybvcorftbyeh5bc7p4e

SLU for Voice Command in Smart Home: Comparison of Pipeline and End-to-End Approaches

Thierry Desot, Francois Portet, Michel Vacher
2019 2019 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU)  
Spoken Language Understanding (SLU) is typically performed through automatic speech recognition (ASR) and natural language understanding (NLU) in a pipeline.  ...  Index Terms-Spoken language understanding, automatic speech recognition, natural language understanding, ambient intelligence, voice-user interface  ...  Benjamin Lecouteux for their support using the PyTorch seq2seq library and Kaldi.  ... 
doi:10.1109/asru46091.2019.9003891 dblp:conf/asru/DesotPV19 fatcat:dymcb2ats5a3jpau2cv3rujvii

Biological Structure and Function Emerge from Scaling Unsupervised Learning to 250 Million Protein Sequences [article]

Alexander Rives, Siddharth Goyal, Joshua Meier, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, Rob Fergus
2019 bioRxiv   pre-print
Learning the natural distribution of evolutionary protein sequence variation is a logical step toward predictive and generative modeling for biology.  ...  In the field of artificial intelligence, a combination of scale in data and model capacity enabled by unsupervised learning has led to major advances in representation learning and statistical generation  ...  Furthermore we express our appreciation toward Ian Peikon for providing valuable and creative feedback on multiple iterations of the manuscript.  ... 
doi:10.1101/622803 fatcat:gpcfv4go4fb4rnhu3fxiwq2nmi

Towards Spoken Medical Prescription Understanding

Ali Can Kocabiyikoglu, Francois Portet, Herve Blanchon, Jean-Marc Babouchkine
2019 2019 International Conference on Speech Technology and Human-Computer Dialogue (SpeD)  
We briefly describe the overall approach and focus on the Natural Language Understanding process which was approached through slot-filling.  ...  Index Terms-natural language understanding; spoken dialogue systems; medical computing; prescription management systems 978-1-7281-0984-8  ...  Tri-CRF from [29] , [35] is an extension of a linear chain Conditional Random Field (CRF).  ... 
doi:10.1109/sped.2019.8906646 dblp:conf/sped/KocabiyikogluPB19 fatcat:ws75oskkhbbedlgfotx6akdpvm

Data2Text Studio: Automated Text Generation from Structured Data

Longxu Dou, Guanghui Qin, Jinpeng Wang, Jin-Ge Yao, Chin-Yew Lin
2018 Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System Demonstrations  
Data2Text Studio is a platform for automated text generation from structured data.  ...  It is equipped with a Semi-HMMs model to extract high-quality templates and corresponding trigger conditions from parallel data automatically, which improves the interactivity and interpretability of the  ...  virtual assistant to identify and read out the essential part of the structured data in natural language to make it easier to understand.  ... 
doi:10.18653/v1/d18-2003 dblp:conf/emnlp/DouQWYL18 fatcat:f7hqr5o3svde3lsjyxvjrytena

Spoken language understanding from unaligned data using discriminative classification models

F. Mairesse, M. Gasic, F. Jurcicek, S. Keizer, B. Thomson, K. Yu, S. Young
2009 2009 IEEE International Conference on Acoustics, Speech and Signal Processing  
While data-driven methods for spoken language understanding reduce maintenance and portability costs compared with handcrafted parsers, the collection of word-level semantic annotations for training remains  ...  Index Termssemantic analysis, spoken dialogue systems, spoken language understanding  ...  parsing that can learn from unaligned semantic trees, thus allowing for faster data collection and dialogue system deployment.  ... 
doi:10.1109/icassp.2009.4960692 dblp:conf/icassp/MairesseGJKTYY09 fatcat:klmusbleybffxoo3nxyxuh34zu

Deep Learning for Dialogue Systems

Yun-Nung Chen, Asli Celikyilmaz, Dilek Hakkani-Tür
2017 Proceedings of ACL 2017, Tutorial Abstracts  
However, how to successfully apply deep learning based approaches to a dialogue system is still challenging.  ...  The advance of deep learning technologies has recently risen the applications of neural models to dialogue modeling.  ...  The RNN-based NLG can learn from unaligned data by jointly optimizing sentence planning and surface realization, and language variation can be easily achieved by sampling from output candidates (Wen et  ... 
doi:10.18653/v1/p17-5004 dblp:conf/acl/ChenCH17 fatcat:eyltpy5guna6bfsczav3vji7x4

A Developmentally-Inspired Examination of Shape versus Texture Bias in Machines [article]

Alexa R. Tartaglini, Wai Keen Vong, Brenden M. Lake
2022 arXiv   pre-print
Across three experiments, we find that deep neural networks exhibit a preference for shape rather than texture when tested under conditions that more closely replicate the developmental procedure.  ...  Early in development, children learn to extend novel category labels to objects with the same shape, a phenomenon known as the shape bias.  ...  Acknowledgments We thank Erin Grant, Pat Little, Pam Osborn Popp, and the anonymous reviewers for their valuable feedback. This work was supported by NIH grant R90DA043849.  ... 
arXiv:2202.08340v2 fatcat:dqkk5ijjvnbcxoxiutc5ad576q

Encoder-decoder with Focus-mechanism for Sequence Labelling Based Spoken Language Understanding [article]

Su Zhu, Kai Yu
2017 arXiv   pre-print
This paper investigates the framework of encoder-decoder with attention for sequence labelling based spoken language understanding.  ...  To address this limitation, we propose a novel focus mechanism for encoder-decoder framework.  ...  random fields (CRFs) [3] , and support vector machines (SVMs) [4] .  ... 
arXiv:1608.02097v2 fatcat:3wj3krwsrzgtpmiacbp6sga3o4

Contradiction Detection with Contradiction-Specific Word Embedding

Luyang Li, Bing Qin, Ting Liu
2017 Algorithms  
Despite the effectiveness of traditional context-based word embedding learning algorithms in many natural language processing tasks, such algorithms are not powerful enough for contradiction detection.  ...  CWE is learned from a training corpus which is automatically generated from the paraphrase database, and is naturally applied as features to carry out contradiction detection in SemEval 2014 benchmark  ...  Word Representation Learning. Word representation is central to natural language processing (NLP).  ... 
doi:10.3390/a10020059 fatcat:tqzwdrfid5e7tos23zdjihdq5q

Review Neural Networks about Image Transformation Based on IGC Learning Framework with Annotated Information [article]

Yuanjie Yan, Suorong Yang, Yan Wang, Jian Zhao, Furao Shen
2022 arXiv   pre-print
Furthermore, experiments have been performed to verify the effectiveness of IGC learning. Finally, new research directions and open problems are discussed for future research.  ...  From the perspective of this framework, we review those subtasks and give a unified interpretation of various scenarios.  ...  In NLP, General Language Understanding Evaluation (GLUE) [11] also summarizes nine datasets for different subtasks about neural language.  ... 
arXiv:2206.10155v1 fatcat:qhlkxu3dzjc7bakdhv776h7x4m

Unnatural Language Processing: Bridging the Gap Between Synthetic and Natural Language Data [article]

Alana Marzoev, Samuel Madden, M. Frans Kaashoek, Michael Cafarella, Jacob Andreas
2020 arXiv   pre-print
To generalize to natural utterances, we automatically find projections of natural language utterances onto the support of the synthetic language, using learned sentence embeddings to define a distance  ...  We address this problem by introducing a general purpose technique for "simulation-to-real" transfer in language understanding problems with a delimited set of target behaviors, making it possible to develop  ...  By leveraging two cheap, readily-available sources of supervision-unaligned natural language text and synthetic language about target tasks-it is possible to build broad-coverage systems for language understanding  ... 
arXiv:2004.13645v1 fatcat:vee34acnwzdhte6e5djherpt6q

Improving Multilingual Frame Identification by Estimating Frame Transferability

Jennifer Sikos, Michael Roth, Sebastian Padó
2022 Linguistic Issues in Language Technology  
We measure success by training and testing frame identification models for the target language.  ...  We find that transferability is indeed a useful indicator and supports a setup where a limited amount of target language data is sufficient to train frame identification systems.  ...  We would like to thank the editors of LiLT for their con-  ... 
doi:10.33011/lilt.v19i.939 fatcat:spojwmdgnvh5fp2fwdcjsqc7uq
« Previous Showing results 1 — 15 out of 872 results