Vivien Mast

Probabilistic reference and grounding with PRAGR for dialogues with robots

In this paper, we present a system for effective referential human–robot communication in the face of perceptual deviation using the Probabilistic Reference And GRounding mechanism PRAGR and vague feature models based on prototypes. PRAGR can handle descriptions of arbitrary complexity including spatial relations and uses flexible concept assignment in generation and resolution of referring expressions for bridging conceptual gaps in referential robot–robot or human–robot interaction. We evaluate the benefit of using vague as compared to crisp properties regarding referential success and robustness towards perspective alignment error in referential robot–robot and human–robot communication.

Boundaries and Prototypes in Categorizing Direction

Projective terms such as left, right, front, back are conceptually interesting due to their flexibility of contextual usage and their central relevance to human spatial cognition. Their default acceptability areas are well known, with prototypical axes representing their most central usage and decreasing acceptability away from the axes. Previous research has shown these axes to be boundaries in certain non-linguistic tasks, indicating an inverse relationship between linguistic and non-linguistic direction concepts under specific circumstances. Given this striking mismatch, our study asks how such inverse non-linguistic concepts are represented in language, as well as how people describe their categorization. Our findings highlight two distinct grouping strategies reminiscent of theories of human categorization: prototype based or boundary based. These lead to different linguistic as well as non-linguistic patterns.

Using Foot-Syllable Grammars to Customize Speech Recognizers for Dialogue Systems

This paper aims at improving the accuracy of user utterance understanding for an intelligent German-speaking wheelchair. We compare three different corpus-based context-free restriction grammars for its speech recognizer, which were tested for surface recognition and semantic feature extraction on a dedicated corpus of 135 utterances collected in an experiment with 13 participants. We show that grammars based on phonologically motivated units such as the foot and the syllable outperform phrase-structure grammars in complex scenarios where the extraction of a large number of semantic features is necessary.