Abteen Ebrahimi, Manuel Mager, Arturo Oncevay, Vishrav Chaudhary, Luis Chiruzzo, Angela Fan, John Ortega, Ricardo Ramos, Annette Rios, Ivan Vladimir Meza Ruiz, Gustavo Giménez-Lugo, Elisabeth Mager, Graham Neubig, Alexis Palmer, Rolando Coto-Solano, Thang Vu, Katharina Kann. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.
Manuel Mager, Arturo Oncevay, Abteen Ebrahimi, John Ortega, Annette Rios, Angela Fan, Ximena Gutierrez-Vasques, Luis Chiruzzo, Gustavo Giménez-Lugo, Ricardo Ramos, Ivan Vladimir Meza Ruiz, Rolando Coto-Solano, Alexis Palmer, Elisabeth Mager-Hois, Vishrav Chaudhary, Graham Neubig, Ngoc Thang Vu, Katharina Kann. Proceedings of the First Workshop on Natural Language Processing for Indigenous Languages of the Americas. 2021.
1 January 2019 in International Journal of Advanced Robotic Systems
The facial expression of angry emotion can be useful to direct the interaction between agents, especially in unclear and cluttered environments. During the presence of an angry face, a process of analysis and diagnosis is activated in the subject that notices it, which could impact its behavior toward the one who expresses the emotion. In order to study such an effect in human–robot interaction, an expressive robotics face was designed and constructed. The influence of this face on human action and attention was analyzed in two collaborative tasks. Results of a digital survey, experimental interaction, and a questionnaire indicated that anger is the best recognized universal facial expression, has a regulatory effect in human action, and induces human attention when an unclear condition arises during the task. An additional finding was that the prolonged presence of an angry face reduces its impact compared to positive expressions.
2 December 2022 in Frontiers in Artificial Intelligence
Little attention has been paid to the development of human language technology for truly low-resource languages—i.e., languages with limited amounts of digitally available text data, such as Indigenous languages. However, it has been shown that pretrained multilingual models are able to perform crosslingual transfer in a zero-shot setting even for low-resource languages which are unseen during pretraining. Yet, prior work evaluating performance on unseen languages has largely been limited to shallow token-level tasks. It remains unclear if zero-shot learning of deeper semantic tasks is possible for unseen languages. To explore this question, we present AmericasNLI, a natural language inference dataset covering 10 Indigenous languages of the Americas. We conduct experiments with pretrained models, exploring zero-shot learning in combination with model adaptation. Furthermore, as AmericasNLI is a multiway parallel dataset, we use it to benchmark the performance of different machine translation models for those languages. Finally, using a standard transformer model, we explore translation-based approaches for natural language inference. We find that the zero-shot performance of pretrained models without adaptation is poor for all languages in AmericasNLI, but model adaptation via continued pretraining results in improvements. All machine translation models are rather weak, but, surprisingly, translation-based approaches to natural language inference outperform all other models on that task.
1 January 2013 in International Journal of Advanced Robotic Systems
In this paper we present SitLog: a declarative situation-oriented logical language for programming situated service robot tasks. The formalism is task and domain independent, and can be used in a wide variety of settings. SitLog can also be seen as a behaviour engineering specification and interpretation formalism to support action selection by autonomous agents during the execution of complex tasks. The language combines the recursive transition network formalism, extended with functions to express dynamic and contextualized task structures, with a functional language to express control and content information. The SitLog interpreter is written in Prolog and SitLog's programs follow closely the Prolog notation, permitting the declarative specification and direct interpretation of complex applications in a modular and compact form. We discuss the structure and representation of service robot tasks in practical settings and how these can be expressed in SitLog. The present framework has been tested in the service robot Golem-II+ using the specification and programming of the typical tasks which require completion in the RoboCup@Home Competition.