<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-0712">
  <Title>Ellipsis Resolution by Controlled Default Unification for Multi-modal and Speech Dialog Systems</Title>
  <Section position="4" start_page="0" end_page="75" type="metho">
    <SectionTitle>
3 Default Unification on Substructures
</SectionTitle>
    <Paragraph position="0"> While classical studies focus on parallism, the importance of non-parallel and fragmentary ellipsis is shown by empirical analysis of spoken dialog (cf.</Paragraph>
    <Paragraph position="1"> (Fernandz and Ginzburg, 2002)). The focus of an elliptic utterance often has no direct counterpart in the antecedent, which makes Rooth's matching condition not directly applicable (cf. (Rooth, 1992), (Hardt and Romero, 2001)). Grammatically required verbs (e.g., the semantically weak verb do) may be omitted in dialog ellipsis. In German spoken language, this is also possible in single and sequential utterances of one speaker.</Paragraph>
    <Paragraph position="2"> We take an example from TALKY, which is the appointment management multimodal dialog system that was developed in the framework of DI-AMOD. The reaction of the system to the first utterance of the user is not necessarily important, because users often proceed with (3) without waiting for the system's answer (i.e., by barge in) or without paying much attention to the system's reaction  (in case of an experienced user).</Paragraph>
    <Paragraph position="3"> 1. USER: Ich m&amp;quot;ochte einen Termin eintragen. (I want to enter an appointment) 2. SYSTEM: presents a new appointment entry 3. USER: mit Schmid (with Schmid)  We achieve the following two representations of  &amp;quot;Matching&amp;quot; cannot be achieved by assuming that there is a hidden attitude connected to very utterance which could be inserted.</Paragraph>
    <Paragraph position="4"> Instead, we search for &amp;quot;matching&amp;quot; nodes with comparable types before normal default unification is applied: thing-with-participant unifies with appointment, which leads to:  In principle, it is quite possible that thing-with-participant describes a certain (collective) type of agents. In this case, the processing would produce an ambiguity. In the DIAMOD system as in many other dialog systems, the agent role is usually restricted to the user and to incarnations of the system. It is not alway posible to find a matching type. In this case we try to find paths that connect terminal nodes of the antecedent structure with the top node of the elliptic structure. It is important, that such connection paths do not introduce new structures corresponding to verbal complements or subordinated sentences.</Paragraph>
    <Paragraph position="5"> If no match is achieved we get simply the new structure back, which is the normal result of applying default unification to non-matching structures.</Paragraph>
  </Section>
  <Section position="5" start_page="75" end_page="75" type="metho">
    <SectionTitle>
4 Task Completion as a Barrier for
</SectionTitle>
    <Paragraph position="0"> Elliptic reference In the following example (taken from Talky), the  user performs her specification in a stepwise manner by extensively using ellipsis. 1. USER: Ich m&amp;quot;ochte am Montag ein Treffen eintragen. (I want to enter a meeting at monday) 2. SYSTEM: Presents an empty appointment entry null 3. USER: Im Bananensaal (In the &amp;quot;banana room&amp;quot;) 4. SYSTEM: Presents appointment entry with banana room 5. USER: Ich meine im Raum Leibniz (I mean in room Leibniz) 6. SYSTEM: Presents appointment entry with room Leibniz 7. USER: um sechs Uhr (at six o'clock) 8. SYSTEM: Presents appointment with room and begin time 6 a.m 9. USER: abends (at the evening) 10. SYSTEM: Presents appointment with room  and begin time 6 p.m Some information has been corrected or clarified, but there was no information removed implicitly. Locally, most steps could be considerd as a case of fragmentary elaboration of the preceding utterances (cf. (Schlangen and Lascarides, 2003)). But this classification depends on more general properties of the dialog. When the task is finished, the availability of old information has changed: 1 USER: Bitte das Treffen jetzt eintragen. (Please enter now the meeting</Paragraph>
  </Section>
  <Section position="6" start_page="75" end_page="75" type="metho">
    <SectionTitle>
2 SYSTEM: Indicates that the meeting is stored
3 USER
</SectionTitle>
    <Paragraph position="0"> a Bitte jetzt ein Treffen am Freitag eintragen please enter a meeting at Friday now b Und am Freitag. And at Friday! c Am Freitag. At Friday! In case of 3a, the the old information is removed. With 3b we recognize that the activity (entering a meeting in a schedule) is still available for being inherited elliptically, while further information, accumulated before, is no longer relevant. If the user wants to keep the more elements of the old information, she has to use anaphoric references, e.g., 4 Und dasselbe am Freitag (And the same at Friday). null The elliptic reading in (3b) is very clear, (3c) is rather an incomplete utterance that has to be clarified. This is also quite different from the specification phase of the meeting.</Paragraph>
    <Paragraph position="1"> Task completion is a barrier for fragmentary elaboration. 4 After task completion, an elliptic relation has to be be marked (e.g., by clue words as und (and). Even then, ellipsis does not refer to the whole information accumulated before, but rather to the utterance that introduced the specification phase of the preceding task.</Paragraph>
  </Section>
  <Section position="7" start_page="75" end_page="75" type="metho">
    <SectionTitle>
5 Information Browsing
</SectionTitle>
    <Paragraph position="0"> Typically, information request are answered after every user input without a lengthly specification phase. As in the case of elliptic specifications, clarification dialog does not affect the elliptic relations between subsequent user queries. If the system actively proposes an action, this will be different.</Paragraph>
    <Paragraph position="1"> Browsing means to vary requests either because it is not clear in advance which information is relevant, how exactly it can be obtained, or because the user wants to gather broad information in some area.</Paragraph>
    <Paragraph position="2"> In browsing dialog, ellipsis is controlled by relations between the informational content of the antecedent and the elliptic utterance. According to our remarks at the beginn of the section, we omit the reactions of the system in the subsequent examples.</Paragraph>
    <Paragraph position="3"> By a group, we understand a collection of information that is orthogonal to other information. By 4The reader may recognize a certain similarity of the considerations in this section with the approach of (Grosz and Sidner, 1986). An example: We restrict ourself to some remarks: Grosz &amp; Sidner focus on the segmentation of discourse along the hierarchical structure of a task, while we focus on problems concerning repetition (this section) and variation of tasks (next section). Grosz &amp; Sidner are mainly concerned with anaphoric reference while we are concerned with ellipsis and related implicit inheritance of information. In our approach, structural relations between information is as much important as aspects concerning the processing of tasks. Furthermore, we discuss problems in relation to a special resolution mechanism, i.e., de- null orthogonal we mean independent and not &amp;quot;competing&amp;quot;. For instance, we consider TIME, LOCATION and CONTENT as basic groups of the information that belongs to a performance. Independence is not a sufficient criterion. Actor and genre are independent, but as our examples may show there are considered as competing. We have no formal means to recognize a group. The knowledge about groups has to be provided.</Paragraph>
    <Paragraph position="4"> We use the term information element (IE) of a feature structure as follows: An IE consists of two parts: a role path and a semantic content. Differing from the usual definition of paths (Carpenter, 1992), a role path is a sequence of alternating types and features (T1F1...TnFn with Types Ti and Features fi). The semantic content is expressed by the sub-structure which is identified by applying the subsequence of the features of the role path (accordingly to standard definition). Role paths can be translated directly in an obvious way into feature structures.</Paragraph>
    <Paragraph position="5"> We speak of an terminal information element (TIE), if the substructure is a type without further specification. A TIE is atomic, if its semantic content is atomic. We represent TIEs as extended role paths by taking the type which expresses their semantic content as last element of the path.</Paragraph>
    <Paragraph position="6"> Two TIEs (or IEs) are of the same sort, if their role path has a common prefix. Two TIEs are of the same terminal sort, if their role paths are identical.</Paragraph>
    <Paragraph position="7"> TIE1 is more general as TIE2 if TIE1 subsumes TIE2. TIE1 subsumes TIE2 if the subsumption relation holds between their translations to feature structures. It will turn out that this definition is to narrow and does not cover the intuitive meaning of being more general.</Paragraph>
    <Paragraph position="8"> The TIEs in elliptic expressions are usually less specific or have a shorter role path than the TIEs in the antecedent. Subsequently we assume that the matching process (as described in former sections) has already been applied and that the TIEs of the elliptic expression are extendend by appending the role path from the root of the antecedent to the matching node. Otherwise we could not correctly determine if an IE is subsumed by another or if they belong to the same group etc.</Paragraph>
    <Paragraph position="9"> We only consider readings of elliptic expressions that amount to a new request, ignoring other readings of elliptic expressions, e.g., as positive or negative feedback.</Paragraph>
  </Section>
  <Section position="8" start_page="75" end_page="75" type="metho">
    <SectionTitle>
3 USER:
</SectionTitle>
    <Paragraph position="0"> a Welche Sience fiction filme laufen? / Und science fiction? / Science fiction! (Which science fiction movies are on? / And science fiction? / science fiction) b Sind Science fiction filme dabei? (Are science fiction movies among them?) In (2) the general information movies (i.e., the TIE informationSearch:TOPIC:performance!in!cinema:CONTENT:movie) is replaced by the coresponding concrete information crime movies (i.e., the TIE informationSearch:TOPIC:performance!in!cinema:CONTENT:- null movie:GENRE:crime). All other information belongs to different groups and is retained. In (3) the information crime movies is replaced by information of the same terminal sort. The specification crime movies is deleted. GENRE is a set-valued feature. Note that set-valued features act quite differently depending on the context (information browsing vs. task specification). If the information crime should be retained, this has to be indicated, e.g. by an anaphorical relation to the result of query (2) as is done in (3b). The reading of (2) and (3) is not affected by the form of the ellipsis, but the strong indication of parallelism that is expressed with &amp;quot;Und Krimis&amp;quot; (&amp;quot;And crime (movies)&amp;quot;) seems not acceptable due to the proper subsumption relation between movies and movies with genre crime.</Paragraph>
    <Paragraph position="1"> 1 USER: Welche Science fiction laufen heute abend in Saarbr&amp;quot;ucken?. (Which science fiction (movies) are on today evening at Saarbr&amp;quot;ucken )</Paragraph>
  </Section>
  <Section position="9" start_page="75" end_page="75" type="metho">
    <SectionTitle>
4 USER:
</SectionTitle>
    <Paragraph position="0"> a Mit Bruce Willis? (With Bruce Willis?) b Und mit Bruce Willis / Welche filme mit Bruce Willis laufen (And with Bruce Willis? / Which movies with Bruce Willis are on) In (4), the new information element Bruce Willis does not belong to the same terminal sort as any element in the antecedent, but by contributing to the specification of movies it belongs to the same group as science fiction. It is a competing element of 'science fiction', and its effect on the information element 'science fiction' is a mixture of the effect of elements of the same sort and elements of a different group (as may be expected). 4b is an an acceptable utterance in this context and it has the effect of deleting the genre information, while 4a without explicit ellipsis indication could also count as adding a specification.</Paragraph>
    <Paragraph position="1"> 1 USER: Welche Krimis kommen heute abend in Saarbr&amp;quot;ucken (Which crime (movies) are on today evening at saarbr&amp;quot;ucken?)</Paragraph>
  </Section>
  <Section position="10" start_page="75" end_page="75" type="metho">
    <SectionTitle>
5 USER:
</SectionTitle>
    <Paragraph position="0"> a Und in Saarlouis (And at Saarlouis?) b Welche filme laufen in Saarlouis? (Which movies are on at Saarlouis?) In (5) the information Saarbr&amp;quot;ucken is replaced by an information of the same terminal sort. 5a has the reading crime movies in Saarlouis. In 5b crime movies is replaced by a more general information. This is an indication that the specification crime should be removed. But Welche filme (which movies) has two other (less preferred) readings: an anaphoric reading which (of those) movies are (also) running at Saarlouis, or even an elliptic (or E-type) reading which crime movies are on at Saarlouis.</Paragraph>
    <Paragraph position="1"> That movie is more general than crime movie can directly inferred from examining the ontology, i.e.</Paragraph>
    <Paragraph position="2"> by subsumption.</Paragraph>
    <Paragraph position="3"> 1 USER: Welche Krimis kommen heute abend in Saarbr&amp;quot;ucken (Which crime (movies) are on?)</Paragraph>
  </Section>
  <Section position="11" start_page="75" end_page="75" type="metho">
    <SectionTitle>
6 USER:
</SectionTitle>
    <Paragraph position="0"> a Und im Scala (And at the Scala (movie theater)) b Welche filme laufen im Scala (Which movies are on at the Scala movie theater) In (6), Saarbr&amp;quot;ucken is replaced by a more concrete information of the sort location. The Scala movie theater is expected to be in Saarbr&amp;quot;ucken except for Scala is a aforementioned cinema in another town. The readings are quite similar to (5). But there is one difference: assume (1) gets an empty result. Than (5a) is still appropriate while (6a) is quite odd. (5b remains (slightly) ambiguous, while (6b) has only one reading. The problem with these findings is, that we cannot recognize by subsumption that Scala is more specific than Saarbr&amp;quot;ucken.</Paragraph>
    <Paragraph position="1"> In information browsing, the relations between the information elements contained in the antecedent and the information elements provided by the ellipsis expression are relevant for resolution. Concrete Information Rule If the elliptic expression contains a more concrete TIE than the antecedent, old specifications that belong to another group are retained.</Paragraph>
    <Paragraph position="2"> General Information Rule If the elliptic expression contains more general information than the antecedent, then the general information tends to be understood as deleting the corresponding concrete Information. The more general TIE introduces a choice points for default unification. Default unification has to produce a reading (usually the more likely one) that accepts general information elements as potential barriers for default unification and removes old information which is beyond the barriers.</Paragraph>
    <Paragraph position="3"> Same Sort Rule If the elliptic expression contains information of the same terminal sort, the old information is deleted, even if the information elements belong to a set-valued feature, except it is made explicit that the feature should be added.</Paragraph>
    <Paragraph position="4"> Competing Information Rule If the elliptic expression indicates parallelism and contains &amp;quot;competing&amp;quot; information of the same group, but not the same terminal sort, the old information is deleted. Otherwise, competing information can be understood as adding a further specification.</Paragraph>
    <Paragraph position="5"> Negative Result Condition Ambiguous readings are sensible for the result of the antecedent query. Negative (empty) results excludes readings that make the specification more concrete.</Paragraph>
    <Paragraph position="6"> We only consider relations between an antecedent query and a subsequent elliptic query. We do not discuss here relations that come into play if a longer history is considered. The examples show, that default unification has to be controlled by relations between information elements.</Paragraph>
  </Section>
class="xml-element"></Paper>