Donna K. Byron’s research while affiliated with Saarland University and other places

... The TRIPS spoken dialog architecture [127], has been used to develop a number of dialog systems over almost a decade on tasks such as emergency response and evacuation planning [128]. The initial implementation handled turn-taking in the standard rigid way, where as the later version featured incremental interpretation and generation and some other features [129]. ...

Reference:

COMPUTATIONAL MODELING OF TURN-TAKING DYNAMICS IN SPOKEN CONVERSATIONS

... Recent cross-linguistic studies (see e.g. Kaiser and Trueswell 2008, Kaiser 2009, Brown-Schmidt et al. 2005, Byron et al. 2008) have also shown that although salience has a clear effect on the choice of referring expressions, various form-specific factors should be taken into consideration in explaining the use of distinct referential forms. A multiple-constraint approach may outperform the uniform dimension (e.g. ...

Reference:

Third person pronoun forms in Estonian in the light of centering theory

... For instance, Kaiser et al. (2009) found that the resolution of pronouns was influenced more by semantic (and less by syntactic) information than the interpretation of reflexives. These studies therefore suggest a complex interaction of syntactic and semantic factors during reference resolution (see also Brown-Schmidt, Byron, & Tanenhaus, 2004. ...

Reference:

Using the visual world paradigm to study language processing: A review and critical evaluation

... If there are event-favoring properties of the context sentence that human participants are sensitive to, it is a tractable task to build automatic classifiers that learn to recognize such properties. This supports the idea that the task of differentiating anaphoric and pleonastic instances of It (Evans, 2001;Boyd et al., 2005;Bergsma and Yarowsky, 2011;Lee et al., 2016;Loáiciga et al., 2017) could potentially improve performance. ...

Reference:

Event versus entity co-reference: Effects of context and form of referring expression

... A.Koller&R.Petrick/SchedulingandPlanningApplicationswoRKshop(SPARK-08)/2008-09-15 6 Example:"Thewhiterabbitsleeps."(2) S:self NP:subj↓ VP:self sleeps V:self N:self rabbit NP:self the N:self white N:self* {sleep(self,subj)} {rabbit(self)} {white(self)} +Knowledgebase Aderivationtoexpress{sleep(e,a)}: S:e NP:a↓ VP:e sleeps V:e N:a rabbit NP:a the N:a white N:a* S:e VP:e sleeps V:e rabbit NP:a the N:a white A.Koller&R.Petrick/SchedulingandPlanningApplicationswoRKshop(SPARK-08)/2008-09-15 7 Sentencegenerationasplanning •Thesentencegenerationproblemcanbetranslatedintoaplanning problem(Koller&Stone,2007) •Actionsaddelementarytreestoaderivation •Aplanmustsatisfythesyntacticandsemanticlinguisticrequirements [add-sleeps(root,a),add-rabbit(subj(root),a),add-white(subj(root),a)] A.Koller&R.Petrick/SchedulingandPlanningApplicationswoRKshop(SPARK-08)/2008-09-15 8 II.GIVEChallenge •"GeneratingInstructionsinVirtualEnvironments"(Kolleretal.,2007) •AchallengefortheNLGcommunity BuildanNLGsystemcapableofproducingnatural languageinstructionstoguideahumanuserin performingsometaskinavirtualenvironment ...

Reference:

Experiences with planning for natural language generation

... Korean does not have long-distance dependency constructions with whphrases as in English. However, previous research, including Kang (199), Lee (2004), etc., showed that long-distance dependency constructions that license traces exist and include relative clauses, topic constructions, and tough predicate constructions. In order to retrieve the meaning of longdistance dependency constructions, the trace information needs to be syntactically represented, and furthermore, semantic binding between the trace and its filler needs to be specified. ...

Reference:

Enriching the Syntactic Annotation of Korean Treebanks for Higher-Level Processing: A Comparative Study of the Penn Korean Treebank and the 21st Sejong …

... Byron et al. [60] devise a Centering of Optimality (COT) system that uses constraints-based evaluation and sorting for antecedent, implemented in Python using the PYCOT package. Lee and Byron [61] demonstrated that the centering theory proposed by Walker et al. (1994) fails to account for the zero and pronoun anaphora. As per [61], both zero and pronouns consider the subject of the preceding phrase as an antecedent. ...

Reference:

A Detailed Study on Anaphora Resolution System for Asian Languages

... A recent tendency to move toward more natural tasks and complex scenes in identification studies has led to new resources. Several corpora have a broader purpose, namely instruction giving in the context of collaborative treasure hunting in three-dimensional virtual worlds (e.g., QUAKE, Byron & Fosler-Lussier, 2006;GIVE-2, Gargett, Garoufi, Koller, & Striegnitz, 2010;and SCARE, Stoia, Shockley, Byron, & Fosler-Lussier, 2008). Here, data consist of whole conversations between partners cooperating on a task, making it difficult to isolate the impact of prior discourse context on the referring expressions used. ...

Reference:

Producing Referring Expressions in Identification Tasks and Route Directions: What’s the Difference?

... speaking or planning an utterance. As in other situated communication tasks (Koller et al., 2010;Smith et al., 2011) the timing of the robot's utterances is important. For fluent interactions the robot needs to monitor the human's actions and changes in the environment and react to them in a timely manner, potentially by interrupting itself or modifying an utterance mid-stream (Clark and Krych, 2004). ...

Reference:

A farewell to arms: Non-verbal communication for non-humanoid robots

... Since cognitive status is closely tied to referring form choice [6], plans generated with our approach will result in shorter language after text realization. The language may sound more natural as well, since humans tend to avoid forming complex referring expressions when possible [39]. We believe our results demonstrate the utility of the modeling cognitive status at the document planning level. ...

Reference:

Givenness Hierarchy Informed Optimal Document Planning for Situated Human-Robot Interaction