Zobrazeno 1 - 10
of 141
pro vyhledávání: '"Gokhan Tur"'
Publikováno v:
ICASSP
Different flavors of transfer learning have shown tremendous impact in advancing research and applications of machine learning. In this work we study the use of a certain family of transfer learning, where the target domain is mapped to the source do
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e3332337a0f877dd9491cb1509661a9c
https://doi.org/10.1109/icassp39728.2021.9413810
https://doi.org/10.1109/icassp39728.2021.9413810
Publikováno v:
CVPR
GuessWhat?! is a two-player visual dialog guessing game where player A asks a sequence of yes/no questions (Questioner) and makes a final guess (Guesser) about a target object in an image, based on answers from player B (Oracle). Based on this dialog
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::97336e4d59e488f1b418f178778c341d
Masked language models have revolutionized natural language processing systems in the past few years. A recently introduced generalization of masked language models called warped language models are trained to be more robust to the types of errors th
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c7316dee736e132fd4314b22f5cf6e7b
Publikováno v:
INTERSPEECH
Automatic speaker verification systems are vulnerable to audio replay attacks which bypass security by replaying recordings of authorized speakers. Replay attack detection (RA) detection systems built upon Residual Neural Networks (ResNet)s have yiel
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::06e91169a4c15898a279d89e209feefe
https://doi.org/10.21437/interspeech.2020-86
https://doi.org/10.21437/interspeech.2020-86
Autor:
Yue Weng, Gokhan Tur, Chandra Khatri, Hugh Williams, Alexandros Papangelis, Runze Wang, Piero Molino, Sai Sumanth Miryala, Huaixiu Zheng, Franziska Bell, Mahdi Namazifar
Publikováno v:
ICASSP
The quality of automatic speech recognition (ASR) is critical to Dialogue Systems as ASR errors propagate to and directly impact downstream tasks such as language understanding (LU). In this paper, we propose multi-task neural approaches to perform c
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e487151ff5ee794db2d8d53f0cdab22d
http://arxiv.org/abs/2002.00750
http://arxiv.org/abs/2002.00750
Autor:
Lei Shu, Zhaleh Feizollahi, Alexandros Papangelis, Hu Xu, Gokhan Tur, Bing Liu, Yi-Chia Wang, Piero Molino
Publikováno v:
EMNLP (Findings)
This work introduces Focused-Variation Network (FVN), a novel model to control language generation. The main problems in previous controlled language generation models range from the difficulty of generating text according to the given attributes, to
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b8f510b39921e58e5723095f8c6c8e0d
https://doi.org/10.18653/v1/2020.findings-emnlp.339
https://doi.org/10.18653/v1/2020.findings-emnlp.339
Publikováno v:
SIGdial
This paper proposes a novel end-to-end architecture for task-oriented dialogue systems. It is based on a simple and practical yet very effective sequence-to-sequence approach, where language understanding and state tracking tasks are modeled jointly
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::54c72663428bbf941cc9e99ba7facf6e
http://arxiv.org/abs/1908.02402
http://arxiv.org/abs/1908.02402
Publikováno v:
SIGdial
We present the first complete attempt at concurrently training conversational agents that communicate only via self-generated language. Using DSTC2 as seed data, we trained natural language understanding (NLU) and generation (NLG) networks for each a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c574b46e2e07b4964f1d2e59bf448857
Publikováno v:
KDD
Smart reply systems have been developed for various messaging platforms. In this paper, we introduce Uber's smart reply system: one-click-chat (OCC), which is a key enhanced feature on top of the Uber in-app chat system. It enables driver-partners to
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fc89f0cf60f3c9d0b596510c1cf84978
Publikováno v:
SLT
We introduce end-to-end neural network based models for simulating users of task-oriented dialogue systems. User simulation in dialogue systems is crucial from two different perspectives: (i) automatic evaluation of different dialogue models, and (ii
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c5c9a283d1700211f0f0fa3d6041f69a
https://doi.org/10.1109/slt.2018.8639652
https://doi.org/10.1109/slt.2018.8639652