File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/87/p87-1022_metho.xml
Size: 14,923 bytes
Last Modified: 2025-10-06 14:12:02
<?xml version="1.0" standalone="yes"?> <Paper uid="P87-1022"> <Title>A CENTERING APPROACH TO PRONOUNS</Title> <Section position="3" start_page="155" end_page="156" type="metho"> <SectionTitle> * CONSTRAINTS </SectionTitle> <Paragraph position="0"> 1. There is precisely one Cb.</Paragraph> <Paragraph position="1"> 2. Every element of Cf(Un) must be realized in U,.</Paragraph> <Paragraph position="2"> 3. Cb(Un) is the highest-ranked element of Cf(U,-1) that is realized in U,.</Paragraph> <Paragraph position="3"> * RULES 1. If some element of Cf(U,-1) is realized as a pronoun in U,, then so is Cb(U,). 2. Continuing is preferred over retaining which is preferred over shifting. As is evident in constraint 3, ranking of the items on the forward center list, Cf, is crucial. We rank the items in Cf by obliqueness of grammatical relation of the subcategorized functions of the main verb: that is, first the subject, object, and object2, followed by other subcategorized functions, and finally, adjuncts. This captures the idea in \[GJW86\] that subjecthood contributes strongly to the priority of an item on the We are aware that this ranking usually coincides with surface constituent order in English. It would be of interest to examine data from languages with relatively freer constituent order (e.g. German) to determine the influence of constituent order upon centering when the grammatical functions are held constant. In addition, languages that provide an identifiable topic function (e.g. Japanese) suggest that topic takes precedence over subject.</Paragraph> <Paragraph position="4"> The part of the HPSG system that uses the centering algorithm for pronoun binding is called the pragmatics processor. It interacts with another module called the semantics processor, which computes representations of intrasentential anaphoric relations, (among other things). The semantics processor has access to information such as the surface syntactic structure of the utterance. It provides the pragmatics processor with representations which include of a set of reference markers. Each reference marker is contraindexed ~ with expressions with which it cannot co-specify 3. Reference markers also carry information about agreement and grammatical function. Each pronominal reference marker has a unique index from Ax,...,An and is displayed in the figures in the form \[POLLARD:A1 L where POLLARD is the semantic representation of the co-specifier. For non-pronominal reference markers the surface string is used as the index. Indices for indefinites are generated from XI,..., X,~.</Paragraph> </Section> <Section position="4" start_page="156" end_page="157" type="metho"> <SectionTitle> 2 Extension </SectionTitle> <Paragraph position="0"> The constraints proposed by \[GJW86\] fail in certain examples like the following (read with pronouns destressed): null Brennan drives an Alfa Romeo.</Paragraph> <Paragraph position="1"> She drives too fast.</Paragraph> <Paragraph position="2"> Friedman races her on weekends.</Paragraph> <Paragraph position="3"> She often beats her.</Paragraph> <Paragraph position="4"> This example is characterized by its multiple ambiguous pronouns and by the fact that the final utterance achieves a shift (see figure 4). A shift is inevitable because of constraint 3, which states that the Cb(U,~) must equal the Cp(U,-I) (since the Cp(Un-x) is directly realized by the subject of Un, &quot;Friedman&quot;). However the constraints and rules from \[GJW86\] would fail to make a choice here between the co-specification possibilities for the pronouns in U,. Given that the transition is a shift, there seem to be more and less coherent ways to shi~. Note that the three items being examined in order to characterize the transition between each pair of anchors 4 are the = See \[BP80\] and \[Cho80\] for conditions on coreference</Paragraph> <Paragraph position="6"> Cb of U,,-1, the Cb of U,~, and the Cp of Un. By \[GJW86\] a shift occurs whenever successive Cb's are not the same. This definition of shifting does not consider whether the Cb of U, and the Cp of Un are equal. It seems that the status of the Cp of Un should be as important in this case as it is in determining the retaining/continuing distinction.</Paragraph> <Paragraph position="7"> Therefore, we propose the following extension which handles some additional cases containing multiple ambiguous pronouns: we have extended rule 2 so that there are two kinds of shifts. A transition for Un is ranked more highly if Cb(Un) = Cp(U,); this state we call shifting-1 and it represents a more coherent way to shift. The preferred ranking is continuing >- retaining >- shifting-1 ~ shifting (see figure 3). This extension enables us to successfully bind the &quot;she&quot; in the final utterance of the example in figure 4 to &quot;Friedman.&quot; The appendix illustrates the application of the algorithm to figure 4.</Paragraph> <Paragraph position="8"> Kameyama \[Kam86\] has proposed another extension to the \[G:JW86\] theory - a property-sharing constraint which attempts to enforce a parallellism between entities in successive utterances. She considers two properties: SUBJ and IDENT. With her extension, subject pronouns prefer subject antecedents and non-subject pronouns prefer non-subject antecedents.</Paragraph> <Paragraph position="9"> However, structural parallelism is a consequence of our ordering the Cf list by grammatical function and the preference for continuing over retaining. Furthermore, the constraints suggested in \[GJW86\] succeed in many cases without invoking an independent structural parallelism constraint, due to the distinction between continuing and retaining, which Kameyama fails to consider. Her example which we reproduce in figure 5 can also be accounted for using the contin- null uing/retaining distinction s. The third utterance in this example has two interpretations which are both consistent with the centering rules and constraints.</Paragraph> <Paragraph position="10"> Because of rule 2, the interpretation in figure 5 is preferred over the one in figure 6.</Paragraph> </Section> <Section position="5" start_page="157" end_page="159" type="metho"> <SectionTitle> 3 Algorithm for centering and </SectionTitle> <Paragraph position="0"> pronoun binding There are three basic phases to this algorithm. First the proposed anchors are constructed, then they are filtered, and finally, they are classified and ranked. The proposed anchors represent all the co-specification relationships available for this utterance. Each step is discussed and illustrated in figure 7.</Paragraph> <Paragraph position="1"> It would be possible to classify and rank the proposed anchors before filtering them without any other changes to the algorithm. In fact, using this strategy I. CONSTRUCT THE PROPOSED ANCHORS for Un (a) Create set of referring expressions (RE's).</Paragraph> <Paragraph position="2"> (b) Order KE's by grammatical relation.</Paragraph> <Paragraph position="3"> (c) Create set of possible forward center (C f) lists. Expand each element of (b) according to whether it is a pronoun or a proper name. Expand pronouns into set with entry for each discourse entity which matches its agreement features and expand proper nouns into a set with an entry for each possible referent. These expansions are a way of encoding a disjunction of possibilities. (d) Create list of possible backward centers (Cb's). This is taken as the entities f~om Cf(U,-1) plus an additional entry of NIL to allow the possibility that we will not find a Cb for the current utterance.</Paragraph> <Paragraph position="4"> (e) Create the proposed anchors. (Cb-O.f combinations from the cross-product of the previous two steps) 2. FILTER THE PROPOSED ANCHORS For each anchor in our list of proposed anchors we apply the following three filters. If it passes each filter then it is still a possible anchor for the current utterance.</Paragraph> <Paragraph position="5"> (a) Filter by contraindices. That is, if we have proposed the same antecedent for two contraindexed pronouns or if we have proposed an antecedent for a pronoun which it is contraindexed with, eliminate this anchor from consideration.</Paragraph> <Paragraph position="6"> (b) Go through Cf(U,_,) keeping (in order) those which appear in the proposed Cf list of the anchor. If the proposed Cb of the anchor does not equal the first element of this constructed list then eliminate this anchor. This guarantees that the Cb will be the highest ranked element of the Cf(U,-t) realized in the current utterance. (This corresponds to constraint 3 given in section t) (c) If none of the entities realized as pronouns in the proposed C\[ list equals the proposed Cb then eliminate this anchor. This guarantees that if any element is realized as a pronoun then the Cb is realized as a pronoun. (If there are no pronouns in the proposed C\[ list then the anchor passes this filter. This corresponds' to rule 1 in section 1). This rule could be implemented as a preference strategy rather than a strict filter.</Paragraph> <Paragraph position="7"> 3. CLASSIFY and BANK EXAMPLE: She doesn't believe him. (U,+4 from figure 2)</Paragraph> <Paragraph position="9"> =~ There are four possible < Cb, Cf > pairs for this utterance.</Paragraph> <Paragraph position="10"> i. <\[POLLARD:A2\], (\['FRIEDMAN:A4\] \[POLLARD:A5\])> ii. <\[FRIEDMAN:A3\], (\[FRIEDMAN:A4\] \[POLLARD:A5\])> iii. <\[KAISE:X1\], (\[FRIEDMAN:A4\] \[POLLARD:A$\])> iv. <NIL, (\[FRIEDMAN:A4\] \[POLLARD:A5\])> =~ This filter doesn't eliminate any of the proposed anchors in this example. Even though \[A4\] and \[A5\] are contraindexed we have not proposed the same co-specifier due to agreement.</Paragraph> <Paragraph position="11"> =~ This filter eliminates proposed anchors ii, iii, iv.</Paragraph> <Paragraph position="12"> =~ This filter doesn't eliminate any of the proposed anchors.</Paragraph> <Paragraph position="13"> The proposed Cb was realized as a pronoun.</Paragraph> <Paragraph position="14"> (a) Classify each anchor on the list of proposed anchors by =~ Anchor i is classified as a retention based on tim transition the transitions as described in section 1 taking U,~-t to state definition. be the previous utterance and U, to be the one we are currently working on.</Paragraph> <Paragraph position="15"> (b) Rank each proposed anchor using the extended rank- =~ Anchor i is the most highly ranked anchor (trivially). ing in section 2. Set Cb(Un) to the proposed Cb and Cf(Un) to proposed Cf of the most highly ranked an null one could see if the highest ranked proposal passed all the filters, or if the next highest did, etc. The three filters in the filtering phase may be done in parallel.</Paragraph> <Paragraph position="16"> The example we use to illustrate the algorithm is in figure 2.</Paragraph> </Section> <Section position="6" start_page="159" end_page="160" type="metho"> <SectionTitle> 4 Discussion </SectionTitle> <Paragraph position="0"/> <Section position="1" start_page="159" end_page="159" type="sub_section"> <SectionTitle> 4.1 Discussion of the algorithm </SectionTitle> <Paragraph position="0"> The goal of the current algorithm design was conceptual clarity rather than efficiency. The hope is that the structure provided will allow easy addition of further constraints and preferences. It would be simple to change the control structure of the algorithm so that it first proposed all the continuing or retaining anchors and then the shifting ones, thus avoiding a precomputation of all possible anchors.</Paragraph> <Paragraph position="1"> \[GJW86\] states that a realization may contribute more than one entity to the Cf(U). This is true in cases when a partially specified semantic description is consistent with more than one interpretation. There is no need to enumerate explicitly all the possible interpretations when constructing possible C f(U)'s 6, as long as the associated semantic theory allows partially specified interpretations. This also holds for entities not directly realized in an utterance. On our view, after referring to &quot;a house&quot; in U,,, a reference to &quot;the door&quot; in U,~+I might be gotten via inference from the representation for '% house&quot; in Cf(Un). Thus when the proposed anchors are constructed there is no possibility of having an infinite number of potential Cf's for an utterance of finite length.</Paragraph> <Paragraph position="2"> Another question is whether the preference ordering of transitions in constraint 3 should always be the same. For some examples, particularly where U,~ contains a single pronoun and U,~-I is a retention, some informants seem to have a preference for shifting, whereas the centering algorithm chooses a continuation (see figure 8). Many of our informants have no strong preference as to the co-specification of the unstressed &quot;She&quot; in Un+4. Speakers can avoid ambiguity by stressing a pronoun with respect to its phonological environment. A computational system for understanding may need to explicitly acknowledge this ambiguity.</Paragraph> <Paragraph position="3"> A computational system for generation would try to plan a retention as a signal of an impending shift, so that after a retention, a shift would be preferred rather than a continuation.</Paragraph> </Section> <Section position="2" start_page="159" end_page="160" type="sub_section"> <SectionTitle> 4.2 Future Research </SectionTitle> <Paragraph position="0"> Of course the local approach described here does not provide all the necessary information for interpreting pronouns; constraints are also imposed by world knowledge, pragmatics, semantics and phonology.</Paragraph> <Paragraph position="1"> There are other interesting questions concerning the centering algorithm. How should the centering algorithm interact with an inferencing mechanism? Should it make choices when there is more than one proposed anchor with the same ranking? In a database query system, how should answers be in- null corporated into the discourse model? How does centering interact with a treatment of definite/indefinite NP's and quantifiers? We are exploring ideas for these and other extensions to the centering approach for modeling reference in local discourse.</Paragraph> </Section> </Section> <Section position="7" start_page="160" end_page="160" type="metho"> <SectionTitle> 5 Acknowledgements </SectionTitle> <Paragraph position="0"> We would like to thank the following people for their help and insight: Hewlett Packard Lab's Natural Language group, CSLI's DIA group, Candy Sidnet, Dan Flickinger, Mark Gawron, :John Nerbonne, Tom Wasow, Barry Arons, Martha Pollack, Aravind :Joshi, two anonymous referees, and especially Barbara Grosz.</Paragraph> </Section> class="xml-element"></Paper>