Ivana Kruijff-Korbayova | German Research Center for Artificial Intelligence (original) (raw)
Uploads
Papers by Ivana Kruijff-Korbayova
2022 31st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)
Proceedings of the CODI-CRAC 2021 Shared Task on Anaphora, Bridging, and Discourse Deixis in Dialogue, 2021
This paper describes the use of Multi-Task Neural Networks (NNs) for system dialogue act selectio... more This paper describes the use of Multi-Task Neural Networks (NNs) for system dialogue act selection. These models leverage the representations learned by the Natural Language Understanding (NLU) unit to enable robust initialization/bootstrapping of dialogue policies from medium sized initial data sets. We evaluate the models on two goal-oriented dialogue corpora in the travel booking domain. Results show the proposed models improve over models trained without knowledge of NLU tasks.
Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue, 2019
Abstract. A synthetic agent requires the coordinated use of multiple sensory and effector modalit... more Abstract. A synthetic agent requires the coordinated use of multiple sensory and effector modalities in order to achieve a social human-robot interaction (HRI). While systems in which such a concatenation of multiple modalities exist, the issue of information coordination across modalities to identify relevant context information remains problematic. A system-wide information formalism is typically used to address the issue, which requires a re-encoding of all information into the system ontology. We propose a general approach to ...
Proceedings of the 9th SIGdial Workshop on Discourse and Dialogue - SIGdial '08, 2008
2019 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), 2019
Mobile robots can provide significant operational advantages in emergency response missions. With... more Mobile robots can provide significant operational advantages in emergency response missions. With increasing autonomy robots need knowledge of the current mission in order to be able to properly contribute to it. We propose to acquire mission knowledge by interpreting the verbal communication among the human response-team members and to use process mining techniques to ground the interpretations in analyses of mission process data and corresponding reference models. We also present a novel concept of mission assistance that uses the acquired mission knowledge to support the first responders' work processes both during and after the mission. The assistance functions include process assistance for the coordination of human-robot team operations; automatic mission documentation generation; and process modeling for first responder training. We describe the architecture of our system and the design and current implementation state of its components: Speech Processing, Mission-Knowledge Management, Process Mining, and Process Assistance. We build on concepts that were evaluated and validated by first responders in a previous project; our extensions have been assessed qualitatively and will be further evaluated in the course of our current project.
2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), 2021
Companion of the 2018 ACM/IEEE International Conference on Human-Robot Interaction, Mar 1, 2018
A World with Robots, 2017
2016 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), 2016
Journal of Human-Robot Interaction, 2015
Lecture Notes in Computer Science, 2015
Proceedings of the Tenth Annual ACM/IEEE International Conference on Human-Robot Interaction Extended Abstracts, 2015
The TRADR project aims at developing methods and models for human-robot teamwork, enabling robots... more The TRADR project aims at developing methods and models for human-robot teamwork, enabling robots to operate in search & rescue environments alongside humans as teammates, rather than as tools. Through a user-centered cognitive engineering method, human-robot teamwork is analyzed, modeled, implemented and evaluated in an iterative fashion. Important is the notion of persistence: rather than treating each sortie as a separate instance for which the build-up of situation awareness and exploration starts from scratch, the objective for the TRADR project is to provide robotic support in an ongoing, fluent manner. This paper provides a short overview of important aspects for human-robot teaming, such as human-robot teamwork coordination and joint situation awareness.
Proceedings of the 12th European Workshop on Natural Language Generation - ENLG '09, 2009
2022 31st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)
Proceedings of the CODI-CRAC 2021 Shared Task on Anaphora, Bridging, and Discourse Deixis in Dialogue, 2021
This paper describes the use of Multi-Task Neural Networks (NNs) for system dialogue act selectio... more This paper describes the use of Multi-Task Neural Networks (NNs) for system dialogue act selection. These models leverage the representations learned by the Natural Language Understanding (NLU) unit to enable robust initialization/bootstrapping of dialogue policies from medium sized initial data sets. We evaluate the models on two goal-oriented dialogue corpora in the travel booking domain. Results show the proposed models improve over models trained without knowledge of NLU tasks.
Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue, 2019
Abstract. A synthetic agent requires the coordinated use of multiple sensory and effector modalit... more Abstract. A synthetic agent requires the coordinated use of multiple sensory and effector modalities in order to achieve a social human-robot interaction (HRI). While systems in which such a concatenation of multiple modalities exist, the issue of information coordination across modalities to identify relevant context information remains problematic. A system-wide information formalism is typically used to address the issue, which requires a re-encoding of all information into the system ontology. We propose a general approach to ...
Proceedings of the 9th SIGdial Workshop on Discourse and Dialogue - SIGdial '08, 2008
2019 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), 2019
Mobile robots can provide significant operational advantages in emergency response missions. With... more Mobile robots can provide significant operational advantages in emergency response missions. With increasing autonomy robots need knowledge of the current mission in order to be able to properly contribute to it. We propose to acquire mission knowledge by interpreting the verbal communication among the human response-team members and to use process mining techniques to ground the interpretations in analyses of mission process data and corresponding reference models. We also present a novel concept of mission assistance that uses the acquired mission knowledge to support the first responders' work processes both during and after the mission. The assistance functions include process assistance for the coordination of human-robot team operations; automatic mission documentation generation; and process modeling for first responder training. We describe the architecture of our system and the design and current implementation state of its components: Speech Processing, Mission-Knowledge Management, Process Mining, and Process Assistance. We build on concepts that were evaluated and validated by first responders in a previous project; our extensions have been assessed qualitatively and will be further evaluated in the course of our current project.
2021 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), 2021
Companion of the 2018 ACM/IEEE International Conference on Human-Robot Interaction, Mar 1, 2018
A World with Robots, 2017
2016 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR), 2016
Journal of Human-Robot Interaction, 2015
Lecture Notes in Computer Science, 2015
Proceedings of the Tenth Annual ACM/IEEE International Conference on Human-Robot Interaction Extended Abstracts, 2015
The TRADR project aims at developing methods and models for human-robot teamwork, enabling robots... more The TRADR project aims at developing methods and models for human-robot teamwork, enabling robots to operate in search & rescue environments alongside humans as teammates, rather than as tools. Through a user-centered cognitive engineering method, human-robot teamwork is analyzed, modeled, implemented and evaluated in an iterative fashion. Important is the notion of persistence: rather than treating each sortie as a separate instance for which the build-up of situation awareness and exploration starts from scratch, the objective for the TRADR project is to provide robotic support in an ongoing, fluent manner. This paper provides a short overview of important aspects for human-robot teaming, such as human-robot teamwork coordination and joint situation awareness.
Proceedings of the 12th European Workshop on Natural Language Generation - ENLG '09, 2009