!DOCTYPE HTML> Workshop -

Workshop on Cognitive Architectures for HRI: Embodied Models of Situated Natural Language Interactions

Click here for the program and the contributions.

The workshop will take place in conjunction with the International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2019) in Montreal, Canada on the 13th of May

In many application fields of human robot interaction, robots need to adapt to changing contexts and thus be able to learn tasks from non-expert humans through verbal and non-verbal interaction. Inspired by human cognition and social interaction, we are interested in mechanisms for representation and acquisition, memory structures etc., up to full models of socially guided, situated, multi-modal language interaction. These models can then be used to test theories of human situated multi-modal interaction, as well as to inform computational models in this area of research.

Call for Papers

The workshop aims at bringing together linguists, computer scientists, cognitive scientists, and psychologists with a particular focus on embodied models of situated natural language interaction. Workshop submissions should answer at least one of the following questions:

  • Which kind of data is adequate to develop socially guided models of language acquisition, e.g. multi-modal interaction data, audio, video, motion tracking, eye tracking, force data (individual or joint object manipulation)?
  • How should empirical data be collected and preprocessed in order to develop socially guided models of language acquisition, e.g. collect either human-human or human-robot data?
  • Which mechanisms are needed by the artificial system to deal with the multi-modal complexity of human interaction. And how to combine information transmitted via different modalities - at a higher level of abstraction?
  • Models of language learning through multi-modal interaction: How should semantic representations or mechanisms for language acquisition look like to allow an extension through multi-modal interaction?
  • Based on the above representations, which machine learning approaches are best suited to handle the multi-modal, time-varying and possibly high dimensional data? How can the system learn incrementally in an open-ended fashion?

Relevant Topics include (but are not limited to) the following:

  • models of embodied language acquisition
  • models of situated natural language interaction
  • multi-modal situated interaction data
  • individual / joint manipulation & task description data
  • multi-modal human-human interaction
  • multi-modal human-robot interaction
  • acquiring multi-modal semantic representations
  • multi-modal reference resolution
  • machine learning approaches for multimodal situated interaction
  • embodied models of incremental learning

Invited Speakers

Keynotes will be given by John Laird, Professor at the faculty of the Computer Science and Engineering Division of the Electrical Engineering and Computer Science Department of the University of Michigan, and Chen Yu, Professor at the Computational Cognition and Learning Lab at Indiana University.

Important Dates

  • Paper submission deadline: February 12 February 22, 2019
  • Notification of acceptance: March 10 March 20, 2019
  • Final version: March 20 March 29, 2019
  • Workshop: May 13 or 14, 2019

Submission Instructions

Articles should be 4-6 pages, formatted using the AAMAS 2019 Author's Kit. For each accepted contribution, at least one of the authors is required to attend the workshop. Authors are invited to submit their manuscripts in PDF here.


Stephanie Gross, Austrian Research Institute for Artificial Intelligence, Vienna, Austria
Brigitte Krenn, Austrian Research Institute for Artificial Intelligence, Vienna, Austria
Matthias Scheutz, Department of Computer Science at Tufts University, Massachusetts, USA
Matthias Hirschmanner, Automation and Control Institute at Vienna University of Technology, Vienna, Austria