Slot Filling & Dialog Systems 2017


Notes:

Approaches already exist allowing conversational slot filling of a structured query within a schema, in the field of spoken dialog systems. Traditional dialog systems can easily inject entities and facts into responses using slot-filling. It has been common practice to employ manually written semantic grammars for slot filling tasks for intent determination. Semantic slot filling is a common practice in dialog systems research that maintains coherence and state by extracting specifics and fitting them into categories. Frame semantics or distributional semantics may be used for slot filling in spoken dialogue systems.

In slot filling applications the complete dialog state is reduced to the state of a small number of slots that require to be filled. Alignment is explicit in slot filling. Position-aware attention and supervised data improve slot filling. Nuance GSL (Grammar Specification Language) utilizes the slot filling concept; for instance, when a grammar rule (in this case made of a clause – a unit of grammatical organization next below the sentence in rank) is hit during speech recognition, variables (slots) are filled with values. 

Knowledge-base population slot filling involves extracting relation information from documents. In slot filling, the task is to complete all known information about a given query entity, for instance search a document collection to fill in values for predefined slots (attributes) for a given entity. So, slots may represent concepts or attributes. Intent detection and slot filling are critical steps for many speech understanding and dialog systems. 

  • Concept annotation
  • Concept annotations
  • Semantic slot
  • Semantic slots
  • Slot annotation
  • Slot annotations

References:

See also:

GSL (Grammar Specification Language)Intent, Dialog Act & Dialog Systems 2017 | Intent, Named Entity & Dialog Systems 2017 | Language Computer Corporation (LCC) | State Tracking & Dialog Systems 2017


End-to-end task-completion neural dialogue systems
X Li, YN Chen, L Li, J Gao – arXiv preprint arXiv:1703.01008, 2017 – arxiv.org
… 2.2 Neural Dialogue System … The popular IOB (in-out-begin) format is used for representing the slot tags, as shown in Figure 2. The LU component is implemented with a single LSTM, which performs intent prediction and slot filling simultaneously (Hakkani-Tür et al., 2016; Chen …

End-to-end joint learning of natural language understanding and dialogue manager
X Yang, YN Chen, D Hakkani-Tür… – … , Speech and Signal …, 2017 – ieeexplore.ieee.org
… Index Terms— language understanding, spoken dialogue systems, end-to-end, dialogue manager, deep learning 1. INTRODUCTION … Traditional approaches for NLU usually model tasks of domain/intent classification and slot filling separately. Se …

Training end-to-end dialogue systems with the ubuntu dialogue corpus
RT Lowe, N Pow, IV Serban, L Charlin… – Dialogue & …, 2017 – dad.uni-bielefeld.de
… This is in contrast to the ‘modular’ system approach to dialogue systems, where each component of Figure 1 is trained separately, and either takes a more structured input, such as a set of dialogue acts, or is trained to maximize an intermediary objective, such as slot-filling …

Spoken language understanding for a nutrition dialogue system
M Korpusik, J Glass – IEEE/ACM Transactions on Audio …, 2017 – ieeexplore.ieee.org
… RELATED WORK To motivate our work on semantic tagging in the food domain, we start by introducing prior work on the similar tasks of spoken language understanding (SLU) in dialogue systems (ie, deter- mining user intent and slot filling, or tagging specific words in a user’s …

Visual dialog
A Das, S Kottur, K Gupta, A Singh… – Proceedings of the …, 2017 – openaccess.thecvf.com
… The former discourages task- engineered bots for ‘slot filling’ [25] and the latter discour- ages bots that put on a personality to avoid answering ques- tions while keeping the user engaged [58 … VisDial Evaluation Protocol One fundamental challenge in dialog systems is evaluation …

Frames: A corpus for adding memory to goal-oriented dialogue systems
LE Asri, H Schulz, S Sharma, J Zumer, J Harris… – arXiv preprint arXiv …, 2017 – arxiv.org
… A generation of voice assistants – such as SIRI, Cortana, and Google Voice – have popularized spoken dialogue systems … assistants, the conversation with a chatbot is very limited: asking for the weather and ordering a cab are accomplished with simple, sequential slot-filling …

A knowledge-grounded neural conversation model
M Ghazvininejad, C Brockett, MW Chang… – arXiv preprint arXiv …, 2017 – arxiv.org
… This contrasts with traditional dialog systems, which can easily inject entities and facts into responses using slot- filling, but often at the cost of significant hand- coding, making such systems difficult to scale to new domains or tasks …

Generative encoder-decoder models for task-oriented spoken dialog systems with chatting capability
T Zhao, A Lu, K Lee, M Eskenazi – arXiv preprint arXiv:1706.08476, 2017 – arxiv.org
… This paper presents a practical and novel framework for building task-oriented dialog systems based on encoder-decoder models … Moreover, this paper shows the flexibility of the proposed method by in- terleaving chatting capability with a slot- filling system for better out-of …

Convolutional neural networks for multi-topic dialog state tracking
H Shi, T Ushio, M Endo, K Yamagami… – Dialogues with Social …, 2017 – Springer
… Footnotes. 1. The original RNN models used in [7] are not designed for the text classification problem, so we did not apply those models to the ‘INFO’ slot filling task. 2 … References. 1. Lemon, O., Pietquin, O.: Data-Driven Methods for Adaptive Spoken Dialogue Systems …

End-to-end optimization of goal-driven and visually grounded dialogue systems
F Strub, H De Vries, J Mary, B Piot, A Courville… – arXiv preprint arXiv …, 2017 – arxiv.org
… dialogue often rely on a predefined structure of the task, such as slot-filling tasks [Williams and Young, 2007] where the task can be casted as filling in a form. In this paper, we present a global architecture for end-to- end RL optimization of a task-oriented dialogue system and its …

Active learning for example-based dialog systems
T Hiraoka, G Neubig, K Yoshino, T Toda… – Dialogues with Social …, 2017 – Springer
… these experiments are available: https://github.com/TakuyaHiraoka/Active-Learning-for-Example- based-Dialog-Systems. 3. We use dialog logs collected from http://www.cleverbot.com/ j2convbydate-page1 … 5. In our research, we use previous dialog act and slot filling status [4 …

Building Task-Oriented Dialogue Systems for Online Shopping.
Z Yan, N Duan, P Chen, M Zhou, J Zhou, Z Li – AAAI, 2017 – aaai.org
… Our proposed approach differs previous work on dialogue system from two aspects: • Training data. Most of previous dialogue system works rely on labeled data as supervision to train statistical models for slot filling, dialog state tracking, police se …

Towards zero-shot frame semantic parsing for domain scaling
A Bapna, G Tur, D Hakkani-Tur, L Heck – arXiv preprint arXiv:1707.02363, 2017 – arxiv.org
… significantly better slot-filling performance when compared to using only in-domain data, especially in the low data regime. Index Terms: slot-filling, deep learning, multi-task RNNs, do- main adaptation, dialogue systems 1. Introduction …

Encoder-decoder with focus-mechanism for sequence labelling based spoken language understanding
S Zhu, K Yu – … , Speech and Signal Processing (ICASSP), 2017 …, 2017 – ieeexplore.ieee.org
… In a spoken dialogue system, the Spoken Language Under- standing (SLU) is a key component that parses user utterances into … The semantic parsing of input utterances in sequence labelling typically consists of three tasks: domain detection, intent determination and slot filling …

Natural Language Does Not Emerge’Naturally’in Multi-Agent Dialog
S Kottur, JMF Moura, S Lee, D Batra – arXiv preprint arXiv:1706.08502, 2017 – arxiv.org
… While historically such agents have been based on slot filling (Lemon et al., 2006), the domi- nant paradigm today is neural dialog models (Bor- des and Weston, 2016; Weston, 2016; Serban et al., 2016a,b) trained on large quantities of data …

Investigation of Language Understanding Impact for Reinforcement Learning Based Dialogue Systems
X Li, YN Chen, L Li, J Gao, A Celikyilmaz – arXiv preprint arXiv:1703.07055, 2017 – arxiv.org
… A traditional dialogue system consists of the following components: 1) a natural language under- standing (NLU) module, which receives utterances of free … frame; usually there are three key tasks in such a NLU mod- ule: domain classification, intent determination and slot filling …

Robust dialog state tracking for large ontologies
F Dernoncourt, JY Lee, TH Bui, HH Bui – Dialogues with Social Robots, 2017 – Springer
… Yet, dialog state tracking is crucial for reliable operations of a spoken dialog system because the latter relies on the estimated dialog state to choose an appropriate response, for example, which tourist … This system is customized for the slot-filling tasks and works as follows …

RUBER: An Unsupervised Method for Automatic Evaluation of Open-Domain Dialog Systems
C Tao, L Mou, D Zhao, R Yan – arXiv preprint arXiv:1701.03079, 2017 – arxiv.org
… Experiments on both re- trieval and generative dialog systems show that RUBER has high correlation with hu- man annotation … In early years, traditional vertical-domain dia- log systems use automatic metrics like slot-filling accuracy and goal-completion rate (Walker et al …

A theoretical framework for conversational search
F Radlinski, N Craswell – Proceedings of the 2017 Conference on …, 2017 – dl.acm.org
… In the field of spoken dialog systems, approaches already exist allowing conversational slot filling of a structured query within a schema (eg [44]). This allows users to book a ticket for a certain concert on a certain night, or set a certain reminder message to …

End-to-End Optimization of Task-Oriented Dialogue Model with Deep Reinforcement Learning
B Liu, G Tur, D Hakkani-Tur, P Shah, L Heck – arXiv preprint arXiv …, 2017 – arxiv.org
… Pomdp-based statistical spoken dialog systems: A review. Proceedings of the IEEE, 2013. [3] Grégoire Mesnil, Yann Dauphin, Kaisheng Yao, Yoshua Bengio, Li Deng, Dilek Hakkani-Tur, et al. Using recurrent neural networks for slot filling in spoken language understanding …

Hybrid dialog state tracker with asr features
M Vodolán, R Kadlec, J Kleindienst – arXiv preprint arXiv:1702.06336, 2017 – arxiv.org
… Abstract This paper presents a hybrid dialog state tracker enhanced by trainable Spoken Language Understanding (SLU) for slot- filling dialog systems. Our architecture is inspired by previously proposed neural- network-based belief-tracking systems …

An end-to-end trainable neural network model with belief tracking for task-oriented dialog
B Liu, I Lane – arXiv preprint arXiv:1708.05956, 2017 – arxiv.org
… [20] B. Liu and I. Lane, “Attention-based recurrent neural network models for joint intent detection and slot filling,” in Proceedings of The … [30] C.-W. Liu, R. Lowe, IV Serban, M. Noseworthy, L. Charlin, and J. Pineau, “How not to evaluate your dialogue system: An empirical study …

Speaker role contextual modeling for language understanding and dialogue policy learning
TC Chi, PC Chen, SY Su, YN Chen – arXiv preprint arXiv:1710.00164, 2017 – arxiv.org
… 2017) introduced network- based end-to-end trainable task-oriented dialogue systems … module— it parses user utterances into semantic frames that capture the core meaning, where three main tasks of LU are domain classification, intent determi- nation, and slot filling (Tur and …

Just ASK: Building an Architecture for Extensible Self-Service Spoken Language Understanding
A Kumar, A Gupta, J Chan, S Tucker… – arXiv preprint arXiv …, 2017 – arxiv.org
… small and sparse datasets and, in doing so, removes significant barriers to entry for software developers and dialog systems researchers … to a particular skill, the Alexa service handles the conversion of speech into text, performs intent classification, and slot-filling according to a …

A simple generative model of incremental reference resolution for situated dialogue
C Kennington, D Schlangen – Computer Speech & Language, 2017 – Elsevier
… early as possible (in other words, time is shared). The overall goal of this paper is to model L’s comprehension process and implement it as a component in a spoken dialogue system. More formally, this can be modelled as a …

Learning conversational systems that interleave task and non-task content
Z Yu, AW Black, AI Rudnicky – arXiv preprint arXiv:1703.00099, 2017 – arxiv.org
… these single-task systems. These single-task systems’ underlying mechanisms are mainly frame-based or agenda-based [Rudnicky and Xu, 1999]. The architecture of traditional dialog systems is slot-filling, which pre-defines the …

Iterative policy learning in end-to-end trainable task-oriented neural dialog models
B Liu, I Lane – arXiv preprint arXiv:1709.06136, 2017 – arxiv.org
… [7] Steve Young, Milica Gašic, Blaise Thomson, and Ja- son D Williams, “Pomdp-based statistical spoken dialog systems: A review … Li Deng, Dilek Hakkani-Tur, Xiaodong He, Larry Heck, Gokhan Tur, Dong Yu, et al., “Using re- current neural networks for slot filling in spoken lan …

A frame tracking model for memory-enhanced dialogue systems
H Schulz, J Zumer, LE Asri, S Sharma – arXiv preprint arXiv:1706.01690, 2017 – arxiv.org
… However, the com- plexity of conversations with current slot-filling dialogue systems is limited. One limitation is that the user usually cannot refer back to an earlier state in the dialogue, which is essential eg, when comparing alternatives or research- ing a complex subject …

Chat Detection in an Intelligent Assistant: Combining Task-oriented and Non-task-oriented Spoken Dialogue Systems
S Akasaki, N Kaji – arXiv preprint arXiv:1705.00746, 2017 – arxiv.org
… Although it lies outside the scope of this paper to explore how to exploit chat detection method in a full dialogue system, the chat detection method is … It is interesting to note that chat detection is not followed by slot-filling unlike intent and domain determination, as far as we use a …

Interactional Dynamics and the Emergence of Language Games
A Eshghi, I Shalyminov, O Lemon – … of the ESSLLI 2017 workshop on …, 2017 – ceur-ws.org
… We show how incremental dialogue systems can be automatically learned from example successful dialogues in a domain, with Dialogue Acts and … This method allows rapid domain trans- fer – simply collect some example (successful) di- alogues in a ‘slot-filling’ domain, and …

Sequential Dialogue Context Modeling for Spoken Language Understanding
A Bapna, G Tur, D Hakkani-Tur, L Heck – Proceedings of the 18th …, 2017 – aclweb.org
… Goal oriented dialogue systems help users with ac- complishing tasks, like making restaurant reserva- tions or booking flights, by interacting with them in … a user utterance is typ- ically broken down into 3 tasks: domain classi- fication, intent classification and slot-filling (Tur and …

Navigation-orientated natural spoken language understanding for intelligent vehicle dialogue
Y Zheng, Y Liu, JHL Hansen – Intelligent Vehicles Symposium …, 2017 – ieeexplore.ieee.org
… Their collaborators at and “HRL” [10] also explored the development of travel and navigation dialogue systems … Xu [3], Guo [19], and Liu [25] proposed DNN architectures for NLP intent detection and slot filling tasks, using Conventional Neural Network (CNN), Recursive Neural …

Towards End-to-End Spoken Dialogue Systems with Turn Embeddings
AO Bayer, EA Stepanov, G Riccardi – Annual Conference of the …, 2017 – sisl.disi.unitn.it
… model jointly with the sequence-to-sequence models improves the performance of the SLU model which is a crucial component of spoken dialogue systems … D. Hakkani- Tur, X. He, L. Heck, G. Tur, D. Yu, and G. Zweig, “Using re- current neural networks for slot filling in spoken …

An Ontology-Based Dialogue Management System for Virtual Personal Assistants
M Wessel, G Acharya, J Carpenter, M Yin – … Spoken Dialogue Systems  …, 2017 – uni-ulm.de
… independent (generic), but dialogue-specific upper-level ontologies and DM rules, which are implementing typical, re-occurring (and usually expensive to address) dialogue system core capabilities, such as anaphora (coreference) resolution, slot- filling, inquiring about missing …

AliMe Assist: An Intelligent Assistant for Creating an Innovative E-commerce Experience
FL Li, M Qiu, H Chen, X Wang, X Gao, J Huang… – Proceedings of the …, 2017 – dl.acm.org
… Typi- cally people use rule- or template- based methods [12], and dialog state tracking [4]. Our slot-filling method is similar to the template- based method. Knowledge-oriented QA … Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models …

Dialog for language to code
S Chaurasia, RJ Mooney – Proceedings of the Eighth International Joint …, 2017 – aclweb.org
… The aim of the dialog system is to determine val- ues of channels and functions for the recipe that the user wants to create. We cast this problem as a slot-filling task, in which the system maintains a belief state — its current estimates for the slots — and follows a hand-coded …

Is spoken language all-or-nothing? Implications for future speech-based human-machine interaction
RK Moore – Dialogues with Social Robots, 2017 – Springer
… 2 Illustration of the consequences of increasing the flexibility of spoken language dialogue systems; increasing flexibility can lead to a … Likewise, simple slot-filling approaches to language understanding and generation are being replaced by sophisticated statistical methods for …

Identifying latent beliefs in customer complaints to trigger epistemic rules for relevant human-bot dialog
C Anantaram, A Sangroya – Control, Automation and Robotics …, 2017 – ieeexplore.ieee.org
… There are recent motivating examples of works that make use of machine-learning to build intelligent dialog systems [5, 4, 6]. Traditional dialog systems are specialized for a domain and rely on slot-filling driven by a knowl edge base and a finite-state model [7, 8]. The finite-state …

Label-dependency coding in Simple Recurrent Networks for Spoken Language Understanding
M Dinarelli, V Vukotic, C Raymond – Interspeech, 2017 – hal.inria.fr
… Index Terms: recurrent neural networks, label dependen- cies,spoken language understanding, slot filling, ATIS, MEDIA … 2.2. MEDIA The research project MEDIA [13] evaluates different SLU mod- els of spoken dialogue systems dedicated to provide tourist in- formation …

A knowledge graph based speech interface for question answering systems
AJ Kumar, C Schmidt, J Köhler – Speech Communication, 2017 – Elsevier
… words, the creation of language models and multi-domain speech recognition, while SLU specific problems include slot filling and intent … on the Web-based distributed computing environment and elaborates how SALT can be used to implement multimodal dialogue systems …

Deep reinforcement learning of dialogue policies with less weight updates
H Cuayáhuitl, S Yu – 2017 – eprints.lincoln.ac.uk
… The elements for training multi- domain DRL-based dialogue systems are as follows … GR is a contribution to slot- filling in the range [-1..1]—it deducts the number of repetitions from the number of slot fillings and confirmations in the cur- rent utterance, and divides by the number …

Dynamic Time-Aware Attention to Speaker Roles and Contexts for Spoken Language Understanding
PC Chen, TC Chi, SY Su, YN Chen – arXiv preprint arXiv:1710.00165, 2017 – arxiv.org
… Contextual information has been incorporated into the recurrent neural network (RNN) for improved domain classification, intent prediction, and slot filling [11, 14, 15, 16]. Most of previous dialogue systems did not take speaker role into consideration …

Test Collections and Measures for Evaluating Customer-Helpdesk Dialogues
Z Zeng, C Luo, L Shang, H Li… – Proceedings of …, 2017 – pdfs.semanticscholar.org
… slot filling schemes that are required by many existing evaluation measures for task-oriented dialogues (See Section 2.2). In the … As an initial step towards evaluating automatic helpdesk dialogue systems, we have constructed a test collection comprising 3,700 real customer …

Challenges for adaptive dialogue management in the KRISTINA project
L Pragst, J Miehle, W Minker, S Ultes – Proceedings of the 1st ACM …, 2017 – dl.acm.org
… Wessel et. al [18] present an ontology-based dialogue manager, utilising slot-filling to build full user requests … A graphical representation can be found in Figure 1. The first step a dialogue system needs to perform is understand- ing what the user has said …

Open-Domain Neural Dialogue Systems
YN Chen, J Gao – Proceedings of the IJCNLP 2017, Tutorial Abstracts, 2017 – aclweb.org
… Although task-oriented dialogue systems and social bots are originally developed for different purposes, there is a trend of combining both … a fully data-driven and knowledge-grounded neural conversation model aimed at producing more contentful responses without slot filling …

Towards a Dialogue System with Long-term, Episodic Memory
D Kondratyuk, C Kennington – … on the Semantics and Pragmatics of …, 2017 – isca-speech.org
… memory in dialogue systems as step towards sys- tems that can better ground conversationally with users … Experiment 1: NLU with Dynamic Memory The goal of this experiment is to determine the usefulness of a slot-filling natural language under- standing system (NLU) using a …

Jointly Modeling Intent Identification and Slot Filling with Contextual and Hierarchical Information
L Wen, X Wang, Z Dong, H Chen – National CCF Conference on Natural …, 2017 – Springer
Intent classification and slot filling are two critical subtasks of natural language understanding (NLU) in task-oriented dialogue systems … Intent classification and slot filling are two critical subtasks of natural language understanding (NLU) in task-oriented dialogue systems …

DialPort: Real-World Data for Academia Spoken Dialog Systems
T Zhao, Y Du, K Lee, M Eskenazi – alborz-geramifard.com
… There are 5 different systems from CMU, including slot-filling systems (eg restaurant, weather, bus etc) and chat-oriented systems. The first external system to connect is the multi-domain dialog system from Cambridge University [5]. The Mr. Clue agent from University of …

Improving Frame Semantic Parsing with Hierarchical Dialogue Encoders
A Bapna, G Túr, D Hakkani-Túr, L Heck – arXiv preprint arXiv:1705.03455, 2017 – arxiv.org
… Goal oriented dialogue systems help users with ac- complishing tasks, like making restaurant reserva- tions or booking flights, by interacting with them in … a user utterance is typ- ically broken down into 3 tasks: domain classi- fication, intent classification and slot-filling (Tur and …

Bootstrapping dialogue systems: the contribution of a semantic model of interactional dynamics
A Eshghi, I Shalyminov, O Lemon – CLASP Papers in Computational …, 2017 – gupea.ub.gu.se
… 6 Conclusion and ongoing work We show how incremental dialogue systems can be automatically bootstrapped from small amounts of successful … This method allows rapid domain transfer–sim- ply collect some example (successful) dialogues in a ‘slot-filling’domain, and retrain …

Utterance Intent Classification of a Spoken Dialogue System with Efficiently Untied Recursive Autoencoders
T Kato, A Nagai, N Noda, R Sumitomo, J Wu… – Proceedings of the 18th …, 2017 – aclweb.org
… system needs to estimate the ut- terance intent correctly despite of various oral ex- pressions. It has been a basic approach to classify the result of automatic speech recognition (ASR) of an utterance into one of multiple predefined in- tent classes, followed with slot filling specific …

Collaboration-based User Simulation for Goal-oriented Dialog Systems
D Didericksen, ORKSL Zhou, J Kramer – alborz-geramifard.com
… Goal-oriented Dialog Systems … The MDP specification (ie, (1)-(3)) is fairly straightforward to define in a goal-oriented slot filling system, making it the workhorse of most CA services, depsite recent research efforts to ameliorate the manual effort of defining slot-based states with …

End-to-End Recurrent Entity Network for Entity-Value Independent Goal-Oriented Dialog Learning
CS Wu, A Madotto, GI Winata, P Fung – workshop.colips.org
… these dialog systems have been built as a pipeline, with modules for language un- derstanding, state tracking, action selection, and language gen- eration [1, 2, 3, 4, 5]. Even though those systems are known to be stable via combining domain-specific knowledge and slot- filling …

Dialogue System for Restaurant Reservations using Hybrid Code Network
C Akin-David, D Xue, E Mei – stanford.edu
… originator of modern task-based dialogue system is the influential GUS system for travel planning (Bobrow et al., 1977). It is a Frame- based Mixed Initiative system. Each frame is a collection of slots and each slot can take certain pre-defined values. Unlike a simple slot-filling …

A study on integrating distinct classifiers with bidirectional LSTM for Slot Filling task
KP Do – 2017 – dspace.jaist.ac.jp
… Page 2. A study on integrating distinct classifiers with bidirectional LSTM for Slot Filling task DO KHAC PHONG School of Information Science … Page 3. Master’s Thesis A study on integrating distinct classifiers with bidirectional LSTM for Slot Filling task 1410220 DO KHAC PHONG …

Dialogue Systems Using Web-based Language Tools
A Niklasson – 2017 – diva-portal.org
… algorithms [3, p. 23]. 2.4 Dialogue management The dialogue manager (DM) has a central role in a dialogue system and coordi … 2.4.2 Frame-based approaches In a frame-based, or slot-filling, approach the state of the dialogue is represented by a frame of information …

LD-SDS: Towards an Expressive Spoken Dialogue System based on Linked-Data
A Papangelis, P Papadakos, M Kotti… – arXiv preprint arXiv …, 2017 – arxiv.org
… for developing such systems, eg: OpenDial [25] is a web-based tool that allows a user to create a slot filling dialogue system automatically, using about fifteen proba- bilistic rules; Olympus is a complete framework for imple- menting spoken dialogue systems (SDS) [4]; PyDial [30 …

Hierarchical Module Classification in Mixed-initiative Conversational Agent System
SXY Suzanna, LL Anthony – Proceedings of the 2017 ACM on …, 2017 – dl.acm.org
… level Information with Encoder LSTM for Semantic Slot Filling. arXiv preprint arXiv:1601.01530 (2016). [9] Ryan Lowe, Nissan Pow, Iulian Serban, and Joelle Pineau. 2015. The ubuntu dialogue corpus: A large dataset for research in unstructured multi-turn dialogue systems …

Using Clockwork RNNs for Dialog State Tracking
K Chandu, A Naik, A Chandrasekar – cs.cmu.edu
… and therefore, comparison of the dialog management approach of these dialog systems is not possible … The architecture of our CW-RNN-based system for dialog state tracking is shown in figure 2. We train one such RNN for each slot-filling task …

Contextual and Structural Language Understanding in Dialogues
YNCD Hakkani-Tür, G Tur, JGACL Deng – pdfs.semanticscholar.org
… Goal-oriented spoken dialogue systems are being incorporated in various devices and allow users to speak to systems freely in order to finish … Traditionally, slot filling is framed as a word sequence tagging task, where the IOB (in-out-begin) format is applied for representing slot …

Exploring ASR-free end-to-end modeling to improve spoken language understanding in a cloud-based dialog system
Y Qian, R Ubale, V Ramanaryanan… – … (ASRU), 2017 IEEE, 2017 – ieeexplore.ieee.org
… [2] V. Ramanarayanan, D. Suendermann-Oeft, AV Ivanov, and K. Evanini, “A distributed cloud-based dialog system for conversational … and G. Zweig, “Using recurrent neural networks for slot filling in spoken language understanding.,” IEEE/ACM Transactions on Audio, Speech …

Improvisational Storytelling Agents
LJ Martin, P Ammanabrolu, X Wang, S Singh… – researchgate.net
… Semantic slot filling is a common practice in dialog systems research that maintains coherence and state by extracting specifics and fitting them into categories (eg. [10]). Our agent framework implements a long-term memory based on event indexing from psychology [11] …

User Intention Classification in an Entities Missed In-vehicle Dialog System
K Zhang, Q Zhu, N Zhang, Z Shi, Y Zhan – International Conference in …, 2017 – Springer
… The control module is based on slot-filling method, which is a typical method used in dialogue systems, and usually has high accuracy; therefore this article will not detail the implementation process of control module. 3 Contextual Intention Classification …

Domain Complexity and Policy Learning in Task-oriented Dialogue Systems
A Papangelis, S Ultes, Y Stylianou – uni-ulm.de
… Task-oriented Dialogue Systems … We formally define information-seeking (or slot-filling) domains (ISD) for dialogue as tuples {S,V,A,D}, where S = {s0,…,sN} is a set of slots, V is a set of values that each slot can take si ? Vi, A is a set of system dialogue acts of the form intent(s0 …

Learning concepts through conversations in spoken dialogue systems
R Jia, L Heck, D Hakkani-Tür… – Acoustics, Speech and …, 2017 – ieeexplore.ieee.org
… [6] Matthew Purver, “CLARIE: Handling clarification requests in a dialogue system,” Research on … Yao, Yoshua Ben- gio, Li Deng, Dilek Hakkani-Tür, Xiaodong He, Larry Heck, Gokhan Tur, Dong Yu, and Geoffrey Zweig, “Using recurrent neural networks for slot filling in spoken …

A Survey of Task-oriented Dialogue Systems
K Mo – 2017 – cse.ust.hk
… Slot-filling: F1 measure Dataset: Air Travel Information System (ATIS) (Dahl et al. 1994), Tourist Information (MEDIA) (Bonneau-Maynard et al. 2005), DARPA Communicator Travel Data (Walker et al. 2001) 11 Page 12. Transfer Learning for Dialogue System …

Extended Hybrid Code Networks for DSTC6 FAIR Dialog Dataset
J Ham, S Lim, KE Kim – workshop.colips.org
… The bot should perform extra tasks to achieve the goal such as asking, querying the database, recommending, and providing additional information. Conventional goal-oriented dialog systems are designed with tracking the dialog state by slot-filling [1], [2], [3], [4]. Young et al …

A Survey on Dialogue Systems: Recent Advances and New Frontiers
H Chen, X Liu, D Yin, J Tang – arXiv preprint arXiv:1711.01731, 2017 – arxiv.org
… Though language understanding is processed by statistical models, most de- ployed dialogue systems still use manual features or hand crafted rules for the state and action space representations, intent detection, and slot filling …

Miscommunication handling in spoken dialog systems based on error-aware dialog state detection
CH Wu, MH Su, WB Liang – EURASIP Journal on Audio, Speech, and …, 2017 – Springer
… December 2017 , 2017:9 | Cite as. Miscommunication handling in spoken dialog systems based on error-aware dialog state detection … Keywords. Error-aware dialog act Miscommunication Spoken dialog systems. Download fulltext PDF. 1 Introduction …

Tracking of enriched dialog states for flexible conversational information access
Y Dai, Z Ou, D Ren, P Yu – arXiv preprint arXiv:1711.03381, 2017 – arxiv.org
… A common practice in current dialog systems is to define the dialog state by a set of slot-value pairs. Such representation of dialog states and the slot- filling based DST have been widely employed, but suffer from three drawbacks …

Deep Learning for Dialogue Systems
YN Chen, A Celikyilmaz, D Hakkani-Tür – Proceedings of ACL 2017 …, 2017 – aclweb.org
… 2007. Partially observable markov decision processes for spoken dialog systems. Computer Speech & Language 21(2):393–422 … Puyang Xu and Ruhi Sarikaya. 2013. Convolutional neural network based triangular CRF for joint intent detection and slot filling …

Learning Generative End-to-end Dialog Systems with Knowledge
T Zhao – 2017 – cs.cmu.edu
… It is sometimes also referred as slot-filling systems. The classical method to build task-oriented system is the pipeline approach, which divides the whole system into several modules as shown in Figure 2.1. The dialog system pipeline contains the following components: natural …

STREAMLInED Challenges: Aligning Research Interests with Shared Tasks
GA Levow, EM Bender, P Littell, K Howell… – Proceedings of the 2nd …, 2017 – aclweb.org
… switching in text and speech. The slot filling task will operate over less-structured human-directed speech, rather than the computer-directed speech prevalent in dialog systems tasks listed above. Finally, the alignment task requires …

Intent detection and semantic parsing for navigation dialogue language processing
Y Zheng, Y Liu, JHL Hansen – … Transportation Systems (ITSC) …, 2017 – ieeexplore.ieee.org
… [8] demonstrated variant joint models for both intent detection and slot filling tasks, using … Another previous work [11] proposed a two-stage framework for the navigation dialogue system, which first utilizes speech recognition to convert the in-vehicle audio stream into text, and …

Roving Mind: a balancing act between open–domain and engaging dialogue systems
A Cervone, G Tortoreto, S Mezza, E Gambi, G Riccardi – researchgate.net
… To the best of our knowledge RM represents the first attempt to build a modular, domain-independent dialogue system architecture, with an explicit … approached as identi- fication of the intent (the purpose behind the utterance) and detection of its arguments (slot-filling) from …

Contextualizing Customer Complaints by Identifying Latent Beliefs and Tailoring a Chatbot’s Dialog through Epistemic Reasoning
C Anantaram, A Sangroya – mrc.kriwi.de
… dialog. On the other hand, traditional dialog systems are specialized for a domain and rely on slot-filling driven by a knowledge base and a finite-state model [Lemon et al., 2006; Wang and Lemon, 2013]. Dynamic epistemic …

Towards a Response Selection System for Spoken Requests in a Physical Domain
A Partovi, I Zukerman, Q Tran – pdfs.semanticscholar.org
… For example, a research proto- type of a spoken slot-filling dialogue system reported a Word Error Rate (WER) of 13.8% when using “a generic dictation ASR system” [Mesnil et al., 2015], and Google reported an 8% WER for its ASR API.1 However, this API had a WER of 54.6 …

Reward-Balancing for Statistical Spoken Dialogue Systems using Multi-objective Reinforcement Learning
S Ultes, P Budzianowski, I Casanueva, N Mrkši?… – arXiv preprint arXiv …, 2017 – arxiv.org
… 40. Oliver Lemon, Kallirroi Georgila, James Henderson, and Matthew Stuttle. 2006. An isu dialogue system Page 6. exhibiting reinforcement learning of dialogue poli- cies: generic slot-filling in the talk in-car system. In Proceedings …

Will this dialogue be unsuccessful? Prediction using audio features
M Kotti, A Papangelis, Y Stylianou – 2017 – scai.info
… Accordingly, the proposed system accomplishes an absolute 17.2% improvement for the case of 13 x 5 MFCC “feature images”. To discuss the limitations of this work, this system refers to a slot-filling dialogue system that works on one specific domain, here Toshiba laptops …

Context-aware selection of multi-modal conversational fillers in human-robot dialogues
M Gallé, E Kynev, N Monet… – Robot and Human …, 2017 – ieeexplore.ieee.org
… distributions at a fixed time k). At the moment when a human finished its utterance h, the different modules of the dialogue system start to … slot filling (depending on the intent, retrieve a pre-defined set of information) and fulfillment (final action, which in this case was just a request …

What’s Up, Doc? A Medical Diagnosis Bot
M Agrawal, J Cheng, C Tran – stanford.edu
… 2 Background There are a number of different paradigms for building dialog systems. According to Allen et … 3.2 Demographic Extraction Slot filling for demographic information, age and gender, is conducted with relatively naive algo- rithms due to the simplicity of the tasks …

Anjishnu Kumar Amazon. com anjikum@ amazon. com
S Tucker, B Hoffmeister, M Dreyer, S Peshterliev… – alborz-geramifard.com
… small and sparse datasets and, in doing so, re- moves significant barriers to entry for software developers and dialogue systems researchers … a particular skill, the Alexa service handles the conversion of speech into text, performs intent classification, and slot-filling according to …

Building Generalize QA System, SLR
M Zoaib, H Raza, H Shabbir, M Suleman, HA Asghar – researchgate.net
… solve another task. These tasks include building QA systems, paraphrase detection, semantic similarity between sentences/words, semantic entailment, machine comprehension, slot filling and other like tasks. We found that …

Let’s Chat about Brexit! A Politically-Sensitive Dialog System Based on Twitter Data
A Khatua, E Cambria, A Khatua… – Data Mining Workshops …, 2017 – ieeexplore.ieee.org
… researchers has also attempted to design a fully data driven knowledge grounded neural conversation model to produce a more content-rich response without slot filling [9]. In … The goal-driven approach relies on hand-crafted rules to generate the response in a dialog system …

Cascaded LSTMs Based Deep Reinforcement Learning for Goal-Driven Dialogue
Y Ma, X Wang, Z Dong, H Chen – National CCF Conference on Natural …, 2017 – Springer
… 1 Introduction. A goal-driven dialogue system usually has three components [1]: Natural Language Understanding (NLU), Dialogue Management (DM), Natural Language Generation (NLG) … The simulators do not set the order of slot filling …

Computer-Based Adaption of Cooking Recipes Integrated in a Speech Dialogue Assistance System
KI Wolf, S Goetze, F Wallhoff – Ambient Assisted Living, 2017 – Springer
… Open image in new window. Fig. 2 Architecture of a speech dialogue system with an optional feedback between dialogue manager and speech recognizer … eg Morbini et al. [13]. For CooCo’s first task, giving recipe advice, a slot-filling system would be sufficient …

Using Context Information for Dialog Act Classification in DNN Framework
Y Liu, K Han, Z Tan, Y Lei – Proceedings of the 2017 Conference on …, 2017 – aclweb.org
… man conversations, as well as for developing intel- ligent human-to-computer dialog systems (either written or spoken dialogs) … For exam- ple, (Rojas-Barahona et al., 2016) proposed to use DNN for DA classification and slot filling, and evaluated on two different sets …

Production Ready Chatbots: Generate if not Retrieve
A Tammewar, M Pamecha, C Jain, A Nagvenkar… – arXiv preprint arXiv …, 2017 – arxiv.org
… define the intent of the state and predefined set of slot filling table which map to responses to gather the necessary entity information … We measure the performance of dialogue systems using percentage of chats that are com- pletely handled by the dialogue system, without any …

A Proposal to Enhance Human-Machine Interaction by Means of Multi-agent Conversational Interfaces
D Griol, AS de Miguel, JM Molina – International Conference on Hybrid …, 2017 – Springer
… The formal description of the proposed model is as follows. Let \(A_{i}\) be the output of the dialog system (the system answer) at time i, expressed in terms of dialog acts … This is the case of most slot-filling conversational systems …

Knowledge Guided Short-Text Classification for Healthcare Applications
S Cao, B Qian, C Yin, X Li, J Wei… – Data Mining (ICDM) …, 2017 – ieeexplore.ieee.org
… 31 Page 2. the intent, as the medical type of “mitral valve prolapse” is a key indicator to the dialog system … Section II provides some relevant background regarding Text Classifica- tion, Name Entity Recognition, Intent Determination and Slot Filling …

Collaborative Response Content Recommendation for Customer Service Agents
C Ma, P Guo, X Xin, X Ma, Y Liang, S Xing, L Li… – … Symposium on Neural …, 2017 – Springer
… Dialogue System: Normally, dialogue system [5] is realized by dialogue management [6], who treats dialogue act (DA) as the input to represent sentence. This type of data is represented using slot-filling. However, our data with …

Concept Transfer Learning for Adaptive Language Understanding
S Zhu, K Yu – arXiv preprint arXiv:1706.00927, 2017 – arxiv.org
… 1 Introduction The language understanding (LU) module is a key component of the dialogue system (DS), pars- ing the users’ utterances into the correspond- ing semantic concepts … Typically, the LU problem is regarded as a slot filling task …

Multimodal Interaction Description Language Based on Data Modeling
M Araki – Multimodal Interaction with W3C Standards, 2017 – Springer
… We proposed a data-modeling driven framework for rapid prototyping of multimodal dialogue systems [6, 7]. The overall architecture of our framework … codes (such as create, list, and edit the instance), controller code (according to task type annotations such as slot-filling type, DB …

FRB-Dialog: A Toolkit for Automatic Learning of Fuzzy-Rule Based (FRB) Dialog Managers
D Griol, AS de Miguel, JM Molina – International Conference on Hybrid …, 2017 – Springer
… Our proposal is focused on slot-filling dialog systems, for which dialog managers use a structure comprised of one slot per piece of information that the system can gather from the user. This data structure, which we call Dialog …

Deep Reinforcement Learning for Inquiry Dialog Policies with Logical Formula Embeddings
T Hiraoka, M Tsuchida, Y Watanabe – arXiv preprint arXiv:1708.00667, 2017 – arxiv.org
… answer questions (or problems) shared by participants [1]. In inquiry dialogs, participants (including the dialog system) do not have complete domain knowledge, so they share their own knowl- edge with their partners. This setting is different from slot- filling dialog settings (eg, [2 …

Rasa: Open Source Language Understanding and Dialogue Management
T Bocklisch, J Faulker, N Pawlowski… – arXiv preprint arXiv …, 2017 – arxiv.org
… As mentioned in the introduction, the large body of high-quality research into statistical dialogue systems from the last decades has not translated into … To demonstrate the usage of Rasa Core, we use the BAbl dialogue dataset [2]. This is a simple slot-filling exercise where the …

Dependency Parsing and Dialogue Systems: an investigation of dependency parsing for commercial application
A Adams – 2017 – diva-portal.org
Page 1. Dependency Parsing and Dialogue Systems an investigation of dependency parsing for commercial application … Page 2. Abstract In this thesis, we investigate dependency parsing for commercial application, namely for future integration in a dialogue system …

” Gimme the Usual”-How Handling of Pragmatics Improves Chatbots
A Bianchini, F Tarasconi, R Ventaglio, M Guadalupi – ceur-ws.org
… Developments in reinforcement learning applications seem promising for task-oriented dialogue systems (Rieser and Lemon, 2011) … Namely: intent classification (eg “booking a flight”); slot filling, ie enriching the intent with more detailed information (such as “destination” and …

Interaction Quality Estimation Using Long Short-Term Memories
N Rach, W Minker, S Ultes – Proceedings of the 18th Annual SIGdial …, 2017 – aclweb.org
… The increasing complexity of Spoken Dialogue Systems (SDS) and the requirements that come with this progress made automatized recognition and … Al- though the herein presented slot filling dialogue is comparatively basic, the IQ is influenced not only by technical aspects (eg …

Dialog for natural language to code
S Chaurasia – 2017 – repositories.lib.utexas.edu
… which can in turn alter the state of the system. In this work, the aim of the dialog system is to determine values of channels and functions for the recipe that the user wants to create. We cast this problem as a slot-filling task: Each piece of information needed to synthesize a …

A Review On Generative Conversational Model
E Varghese, MTR Pillai – data.conferenceworld.in
… users. In this system, it is using deep learning based dialogue system and provide domain specific answers for user queries … (2015) Using recurrent neural networks for slot filling in spoken language understanding. IEEE/ACM Trans Speech Audio Proc 23(3):530–539 …

General Pipeline Architecture for Domain-Specific Dialogue Extraction from different IRC Channels
A Abouzeid – 2017 – content.grin.com
… 11 2.3 Slot Filling in Frame-Based Dialogue Systems . . . . . 12 … Figure 2.3: Slot Filling in Frame-Based Dialogue Systems Agent-Based Systems are Artificial Intelligence-Based designs where data can play a very important role …

Ask Me Otherwise: Synonym-Based Memory Networks for Reading Comprehension
B Srivatsan – bharathsrivatsan.com
Page 1. – Independent Work Report, Spring 2017 – Ask Me Otherwise: Synonym-Based Memory Networks for Reading Comprehension Bharath Srivatsan Advisor: Christiane Fellbaum Abstract Machine learning models built to read and comprehend text are becoming ever more …

Multimodal Fusion and Fission within the W3C MMI Architectural Pattern
D Schnelle-Walka, C Duarte, S Radomski – Multimodal Interaction with …, 2017 – Springer
… 19.1). Open image in new window. Fig. 19.1 High-level architecture of a multimodal dialog system and available W3C standards [24] … They range from well-known slot filling, over unification, to more modern statistical approaches …

Bootstrapping Chatbots for Novel Domains
P Babkin, MFM Chowdhury, A Gliozzo… – Workshop at NIPS on …, 2017 – hirzels.com
… While the resulting dialogue system is able to classify user utterances with reasonable accuracy, it only understands a limited vocabulary and forces the user to use stylized language where the … Using recurrent neural networks for slot filling in spoken language understanding …

Real-time On-Demand Crowd-powered Entity Extraction
THK Huang, YN Chen, JP Bigham – In Proceedings of the 5th Edition …, 2017 – cs.cmu.edu
… 2014. Leveraging semantic web search and browse sessions for multi-turn spoken dialog systems. In ICASSP 2014. IEEE, 4082–4086 … IEEE, 73–78. Puyang Xu and Ruhi Sarikaya. 2014. Targeted Feature Dropout for Robust Slot Filling in Natural Language Understanding …

Natural language understanding and communication for human-robot collaboration
MI Bloch – ipvs.informatik.uni-stuttgart.de
… semantic parsing. Considering autonomously working robots with planning abilities, a dialog system makes these robots to co-workers instead of subordinates, as a dialog system enables these robots to suggest tasks. [Tho …

First Time Encounters with Roberta: a Humanoid Assistant for Conversational Autobiography Creation
M Lee, S Schlögl, S Montenegro, A López, A Ratni… – researchgate.net
… A set of rudimentary dialog agents are found at terminal points of the tree. The main idea behind RavenClaw is simple “slot-filling” … Execute Agent: calls back-end operations, which allows for an open dialog system to handle tasks that are higher in complexity …

Real-time On-Demand Crowd-powered Entity Extraction
YN Chen, JP Bigham – arXiv preprint arXiv:1704.03627, 2017 – arxiv.org
… 2014. Leveraging semantic web search and browse sessions for multi-turn spoken dialog systems. In ICASSP 2014. IEEE, 4082–4086 … IEEE, 73–78. Puyang Xu and Ruhi Sarikaya. 2014. Targeted Feature Dropout for Robust Slot Filling in Natural Language Understanding …

Neural Models for Sequence Chunking.
F Zhai, S Potdar, B Xiang, B Zhou – AAAI, 2017 – aaai.org
… Experimental re- sults show that the proposed neural sequence chunking mod- els can achieve start-of-the-art performance on both the text chunking and slot filling tasks … In addition, it also achieves state-of-the-art performance on both text chunking and slot filling tasks …

May I take your order? A Neural Model for Extracting Structured Information from Conversations
B Peng, M Seltzer, YC Ju, G Zweig… – Proceedings of the 15th …, 2017 – aclweb.org
Page 1. Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 450–459, Valencia, Spain, April 3-7, 2017. cO2017 Association for Computational Linguistics May I take your order …

Transforming unstructured natural language descriptions into measurable process performance indicators using Hidden Markov Models
H van der Aa, H Leopold, A Del-Río-Ortega… – Information Systems, 2017 – Elsevier
… Template filling, which is also referred to as slot filling [30] or semantic-based understanding [31], has been extensively studied and … is spoken language understanding, where information is extracted from unstructured natural language text in the context of a dialog system [32] …

Separating Representation, Reasoning, and Implementation for Interaction Management: Lessons from Automated Planning
ME Foster, RPA Petrick – Dialogues with Social Robots, 2017 – Springer
… For example, the information state might include external aspects such as variables and their assignments (as in a slot-filling dialogue), or it might include internal agent … A similar ISU approach has also been taken in more recent dialogue systems, but using other infrastructure …

Incomplete Follow-up Question Resolution using Retrieval based Sequence to Sequence Learning
V Kumar, S Joshi – Proceedings of the 40th International ACM SIGIR …, 2017 – dl.acm.org
… e rest of the paper is organized as follows. In Section 2, we dis- cuss related work in interactive QA systems, dialogue systems and sequence to sequence learning … 2.2 Dialogue Systems In dialogue systems, context is important for spoken language understanding (SLU) [47] …

Reconciling Event-Based Knowledge Through RDF2VEC
M Alam, DR Recupero, M Mongiovi, A Gangemi… – ceur-ws.org
… [5] apply Frame Semantics and Distributional Semantics for slot filling in Spoken Dialogue System. In [27], the authors use Word and Frame Embeddings for generating categories of annoying behaviors where each category contains a set of words specific to that category …

Improving the understanding of spoken referring expressions through syntactic-semantic and contextual-phonetic error-correction
I Zukerman, A Partovi – Computer Speech & Language, 2017 – Elsevier
… For example, a research prototype of a spoken slot-filling dialogue system reported a Word Error Rate (WER) of 13.8% when using “a generic dictation ASR system” (Mesnil et al., 2015), and Google reported an 8% WER for its ASR API, 1 but this API had a WER of 54.6% when …

Addressee and Response Selection in Multi-Party Conversations with Speaker Interaction RNNs
R Zhang, H Lee, L Polymenakos, D Radev – arXiv preprint arXiv …, 2017 – arxiv.org
… The task requires modeling multi-party conversations and can be directly used to build retrieval- based dialog systems (Lu and Li 2013; Hu et al. 2014; Ji, Lu, and Li 2014; Wang et al. 2015) … 2 Related Work We follow a data-driven approach to dialog systems. Singh et al …

Lexical Acquisition through Implicit Confirmations over Multiple Dialogues
K Ono, R Takeda, E Nichols, M Nakano… – Proceedings of the 18th …, 2017 – aclweb.org
… 1 Introduction Much attention has recently been paid to non-task-oriented dialogue systems —or chat- oriented dialogue systems— both in research (Higashinakaetal., 2014; Yuetal., 2016) and in industry. In addition to pure …

Iris: A Conversational Agent for Complex Tasks
E Fast, B Chen, J Mendelsohn, J Bassen… – arXiv preprint arXiv …, 2017 – arxiv.org
… filling systems [31]), which unstructured systems do not require [27]. Iris is a modular, structured system in that it is trained for a concrete set of tasks and contains structural in- formation provided by its conversational type system. To our knowledge, Iris is the first dialogue system …

Utilizing bots in delivering content from Kentico Cloud and Kentico EMS
A Eikonsalo – 2017 – tampub.uta.fi
… subtler aspects of the user’s utterances. According to Klüwer, dialog systems also generally include some output … tions of the end-user. The flow of the dialog can be defined by configuring con- texts, prioritizing intents, slot filling, responsibilities, and fulfillment by using …

Joint, incremental disfluency detection and utterance segmentation from speech
J Hough, D Schlangen – Proceedings of the 15th Conference of the …, 2017 – aclweb.org
… Joint, Incremental Disfluency Detection and Utterance Segmentation from Speech Julian Hough and David Schlangen Dialogue Systems Group // CITEC // Faculty of Linguistics and Literature Bielefeld University firstname.lastname@uni-bielefeld.de Abstract …

Reinforcement Learning Based Conversational Search Assistant
M Aggarwal, A Arora, S Sodhani… – arXiv preprint arXiv …, 2017 – arxiv.org
… when booking a restaurant, the user may specify preferences in terms of distance and budget which makes it more of a slot filling exercise … There have been various instances of such conversational agents, as spoken dialogue systems [4, 13, 17] or text-based chat bots [5, 7, 8 …

Efficient natural language response suggestion for smart reply
M Henderson, R Al-Rfou, B Strope, Y Sung… – arXiv preprint arXiv …, 2017 – arxiv.org
… For example, neural network models have been used to learn more robust parsers [14, 24, 29]. In recent work, the components of task-oriented dialog systems have been implemented as neural networks, enabling joint learning of robust models [7, 26, 27] …

Domain Transfer for Deep Natural Language Generation from Abstract Meaning Representations
N Dethlefs – IEEE Computational Intelligence Magazine, 2017 – ieeexplore.ieee.org
… This is con- firmed in work by Jaech et al. [44], where the authors showed that training a multi-task model for slot filling in language understanding from several domains gives significantly better performance than training models for single domains …

Symbol sequence search from telephone conversation
M Suzuki, G Kurata, A Sethy… – Proc. Interspeech …, 2017 – isca-speech.org
… There are already many successful applications of spoken interaction, such as IoT applications, dictation, voice search, and spoken dialog systems … B. Xiang, B. Zhou, and M. Yu, “Leveraging sentence- level information with encoder lstm for semantic slot filling,” arXiv preprint …

Effective Spoken Language Labeling with Deep Recurrent Neural Networks
M Dinarelli, Y Dupont, I Tellier – arXiv preprint arXiv:1706.06896, 2017 – arxiv.org
… In this paper, we focus on Spo- ken Language Understanding (SLU), the module of spoken dialog systems responsible for extracting a semantic interpretation from the user utterance. The task is treated as a labeling problem …

19 Speech Synthesis: State of the Art and Challenges for the Future
K Georgila – Social Signal Processing, 2017 – books.google.com
… An ISU dialogue system exhibiting reinforcement learning of dialogue policies: Generic slot-filling in the TALK in-car system. In Proceedings of the Conference of the European Chapter of the Association for Computational Linguistics (EACL)–Demonstrations (pp. 119–122) …

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
M Palmer, R Hwa, S Riedel – Proceedings of the 2017 Conference on …, 2017 – aclweb.org
… 24 Position-aware Attention and Supervised Data Improve Slot Filling Yuhao Zhang, Victor Zhong, Danqi Chen, Gabor Angeli and Christopher D. Manning . . . . . 35 Heterogeneous …

A Hybrid Architecture for Multi-Party Conversational Systems
MG de Bayser, P Cavalin, R Souza, A Braz… – arXiv preprint arXiv …, 2017 – arxiv.org
… initiative in a question and answer mode, while the one in Table 3 is a natural dialogue system where both … understanding (SLU) solution capable of handling multiple context sensitive clas- sification (intent determination) and sequence labeling (slot filling) tasks simultaneously …

Evaluation of Modern Tools for an OMSCS Advisor Chatbot
E Gregori – 2017 – smartech.gatech.edu
… VPINO is a coaching/counseling chatbot based on IBM’s Watson. “?A text based natural language dialogue system specifically developed for the purpose of holding structured, goal directed coaching conversations. VPINO …

Event-based knowledge reconciliation using frame embeddings and frame similarity
M Alam, DR Recupero, M Mongiovi, A Gangemi… – Knowledge-Based …, 2017 – Elsevier
… [33,34] apply Frame Semantics and Distributional Semantics for slot filling in Spoken Dialogue System. In [35], the authors use Word and Frame Embeddings for generating categories of annoying behaviors where each category contains a set of words specific to that category …

Generating chat bots from web API specifications
M Vaziri, L Mandel, A Shinnar, J Siméon… – Proceedings of the 2017 …, 2017 – dl.acm.org
Page 1. Generating Chat Bots from Web API Specifications Mandana Vaziri IBM Research, USA mvaziri@us.ibm.com Louis Mandel IBM Research, USA lmandel@us.ibm.com Avraham Shinnar IBM Research, USA shinnar@us.ibm.com …

End-to-End Trainable Chatbot for Restaurant Recommendations
A Strigér – 2017 – diva-portal.org
… One way to model conversations is as partially observable Markov de- cision processes (POMDPs) which have been used in spoken dialog systems [34, 36] … Another common model is slot-filling [14, 30] where a set of infor- mation slots are predefined …

Deep reinforcement learning: An overview
Y Li – arXiv preprint arXiv:1701.07274, 2017 – arxiv.org
… Then we discuss various applications of RL, including games, in particular, AlphaGo, robotics, spoken dialogue systems (aka chatbot), machine translation, text sequence prediction, neural architecture design, personalized web services, healthcare, finance, and music …

Topic Identification for Speech without ASR
C Liu, J Trmal, M Wiesner, C Harman… – arXiv preprint arXiv …, 2017 – arxiv.org
… ICML, 2008. [13] P. Xu and R. Sarikaya, “Convolutional neural network based tri- angular CRF for joint intent detection and slot filling,” in Proc. ASRU, 2013. [14] C. Liu, P. Xu, and R. Sarikaya, “Deep contextual language under- standing in spoken dialogue systems.” in Proc …

Linguistic Knowledge Transfer for Enriching Vector Representations
JK Kim – 2017 – rave.ohiolink.edu
… sequence tagging tasks. We introduce a cross-domain transfer learning model for dialog slot-filling, which is an inductive transfer learning method, and a cross-lingual transfer … 6.1 The architecture of cross-domain transfer learning model for slot-filling. . . 72 …

CCG Supertagging via Bidirectional LSTM-CRF Neural Architecture
R Kadari, Y Zhang, W Zhang, T Liu – Neurocomputing, 2017 – Elsevier
Sequence labeling is the widely used method for CCG supertagging task where a supertag (lexical category) is assigned to each word in an input sentence. In CCG.

Label-dependencies aware recurrent neural networks
Y Dupont, M Dinarelli, I Tellier – arXiv preprint arXiv:1706.01740, 2017 – arxiv.org
… word 10 Page 11. The ATIS corpus (Air Travel Information System) [26] was collected for building a spoken dialog system able to provide flight information in the United States. ATIS is a simple task dating from 1993. Training …

Naturalizing a programming language via interactive learning
SI Wang, S Ginn, P Liang, CD Manning – arXiv preprint arXiv:1704.06956, 2017 – arxiv.org
Page 1. Naturalizing a Programming Language via Interactive Learning Sida I. Wang, Samuel Ginn, Percy Liang, Christopher D. Manning Computer Science Department Stanford University {sidaw, samginn, pliang, manning}@cs.stanford.edu Abstract …

Situated Intelligent Interactive Systems
Z Yu – 2017 – lti.cs.cmu.edu
… Abstract The recent wide usage of Interactive Systems (or Dialog Systems), such as Apple’s Siri has at- tracted a lot of attention. Besides personal assistants, dialog systems can also be applied in various domains, such as education, health care and entertainment …

End-to-End Architectures for Speech Recognition
Y Miao, F Metze – New Era for Robust Speech Recognition, 2017 – Springer
Automatic speech recognition (ASR) has traditionally integrated ideas from many different domains, such as signal processing (mel-frequency cepstral coefficient features), natural language processing.

Propositional Knowledge: Acquisition and Application to Syntactic and Semantic Parsing
B Cabaleiro Barciela – 2017 – e-spacio.uned.es
… 67 3.4.2 Results . . . . . 68 3.5 Extrinsic Evaluation on Information Extraction . . . . . 69 3.5.1 The Temporal Slot Filling Task . . . . . 69 3.5.2 Experimental Design . . . . . 71 3.5.3 Results …

Learning Semantic Patterns for Question Generation and Question Answering
HP Rodrigues – 2017 – pdfs.semanticscholar.org
… This approach of learning by analogy, or example-based systems [Aamodt and Plaza, 1994], has also been applied in other domains, such as in creation of dialog systems [Nio et al., 2014] or translation of unknown words [Langlais and Patry, 2007] …

Towards the Implementation of an Intelligent Software Agent for the Elderly
AHF Dinevari – 2017 – era.library.ualberta.ca
Page 1. Towards the Implementation of an Intelligent Software Agent for the Elderly by Amir Hossein Faghih Dinevari A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Computing Science University of Alberta …

Advances in Neural Networks-ISNN 2017: 14th International Symposium, ISNN 2017, Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017 …
F Cong, A Leung, Q Wei – 2017 – books.google.com
Page 1. Fengyu Cong· Andrew Leung Qinglai Wei (Eds.) Advances in Neural Networks – ISNN 2017 14th International Symposium, ISNN 2017 Sapporo, Hakodate, and Muroran, Hokkaido, Japan, June 21–26, 2017 Proceedings, Part I 123 Page 2 …

Deep Learning and Reward Design for Reinforcement Learning
X Guo – 2017 – deepblue.lib.umich.edu
… A large and diverse set of problems can be modeled using the MDP framework, such as robot planning and navigation, dialog systems, power supply management, human-computer interaction design and digital marketing …

Refining Word Embeddings Using Intensity Scores for Sentiment Analysis
LC Yu, J Wang, KR Lai, X Zhang – researchgate.net
… [44] proposed a counter-fitting method that injected both antonymy and synonymy relations into vector representations to improve the capability of dialog systems for distinguishing between semantically different but conceptually related words (eg, cheaper and pricey) …

A New Classification Framework to Evaluate the Entity Profiling on the Web: Past, Present and Future
AA Barforoush, H Shirazi, H Emami – ACM Computing Surveys (CSUR), 2017 – dl.acm.org
Page 1. 39 A New Classification Framework to Evaluate the Entity Profiling on the Web: Past, Present and Future AHMAD ABDOLLAHZADEH BARFOROUSH, HOSSEIN SHIRAZI, and HOJJAT EMAMI, Malek Ashtar University of Technology …

Advances in Statistical Script Learning
K Erk – cs.utexas.edu
Page 1. Copyright by Karl Pichotta 2017 Page 2. The Dissertation Committee for Karl Pichotta certifies that this is the approved version of the following dissertation: Advances in Statistical Script Learning Committee: Raymond J. Mooney, Supervisor Nathanael Chambers …

Advances in statistical script learning
K Pichotta – 2017 – repositories.lib.utexas.edu
Page 1. Copyright by Karl Pichotta 2017 Page 2. The Dissertation Committee for Karl Pichotta certifies that this is the approved version of the following dissertation: Advances in Statistical Script Learning Committee: Raymond J. Mooney, Supervisor Nathanael Chambers …

(Visited 79 times, 1 visits today)