<?xml version="1.0" standalone="yes"?>
<Paper uid="W96-0410">
  <Title>Paying Heed to Collocations</Title>
  <Section position="4" start_page="91" end_page="92" type="metho">
    <SectionTitle>
COMBINING EXPRESSIONS (Nunberg et al., 1994)
</SectionTitle>
    <Paragraph position="0"> must be derived compositionally from special, idiomatic meanings of their parts, as when strings = influence, pull = exert privately (from the OED): (2) The strings she pulled didn't get her the job.</Paragraph>
    <Paragraph position="1"> Second, COLLOCATIONS PROPER involve constituents whose meaning is determined by ordinary principles, like copy area, but which must be regarded as conventional in light of the oddness of near synonyms (like duplication zone); such collocations are the subject of the Lexical Functions of the Meaning-Text Theory (MTT) (Mel'~uk and Polgu~re, 1987). Finally, SEMANTIC COLLOCATIONS like long book derive their particular meaning from the recovery in context of parameters for events and other entities (Pustejovsky, 1991).</Paragraph>
    <Paragraph position="2"> Researchers in generation rarely address all of these kinds of conventionality. For example, (Viegas and Bouillon, 1994) handle semantic collocations by implementing Pustejovsky's Generative Lexicon Theory (GLT); modifiers take on specialized meanings derived from salient processes and characteristics associated with the heads they modify. Thus, a long book means a long book to read because of a lexicographic association between books and reading. Similarly, implementations of MTY describe the conventional use of certain modifiers with heads (Mel'~uk and Polgu~re, 1987; Iordanskaja et al., 1991; Wanner, 1994) using Lexical Func- null tions. Thus, a function Magn determines the realization of a concept very, intense, intensely: (3) A Magn escape ~ a narrow escape;  to Magn bleed ~ to bleed profusely.</Paragraph>
    <Paragraph position="3"> Copy area would be handled using the Lexical Function SIoc, which returns the name of the location associated with an activity. (Smadja and McKeown, 1991) are an exception in treating a wide range of conventionality, but they simply list the idiomatic status and meaning of a variety of forms in a way that collapses the diistinct theoretical status, and to a large extent, the distinct meanings, of different collocations.</Paragraph>
    <Paragraph position="4"> These various existing computational approaches have three main deficiencies. First, they derive conventionality from relational lexicons that describe only the properties of WORDS. However, the features that determine appropriateness of conventional attributions are better modelled as properties of OBJECTS in an evolving model of discourse. Idiomatically combining expressions introduce entities for subsequent reference: (4) Kim's family pulled some strings on her behalf, but they weren't enough to get her the job. \[=(Nunberg et al., 1994) 10c\] Semantic collocations recover their parameters based simply on the things described, regardless of their syntactic proximity, as the examples in  (5) show: (5) a I will not check out a long book.</Paragraph>
    <Paragraph position="5">  b I won't check out that book. It's long.</Paragraph>
    <Paragraph position="6"> c I won't check that out. It's a long monstrosity.</Paragraph>
    <Paragraph position="7"> The modifications achieved by Lexical Functions are parallel: as with narrow in (6):  (6) a They made a narrow escape.</Paragraph>
    <Paragraph position="8"> b Their escape had been lucky; Bill found it uncomfortably narrow.</Paragraph>
    <Paragraph position="9"> c Whew! \[after burrowing and swimming out of Alcatraz, amid nearby shots and  searchlights\] That was narrow! Second, by treating different conventional combinations as mere paraphrases of one another, researchers complicate the statement of when and why to use conventional forms. No specification of idiomatic combination is complete without representing the pragmatic circumstances in which its use is appropriate (e.g. saying to someone Your goose is cooked is not appropriate as a expression of sympathy; the expression conveys a certain amount of disregard for their predicament). Meanwhile, some representation of entities and their salience is required to determine whether ellipsis is possible in context. Whether a hard idea is hard to formalize, to communicate, or to understand depends on the topic; to be clear, a natural language system must model how its audience arrives at such understandings.</Paragraph>
    <Paragraph position="10"> Third, by recognizing collocations only when transducing underlying semantic representations, researchers limit the extent to which knowledge of collocations can be exploited in generating flu- null ent text. In particular, transduction presupposes that the content of referring expressions has already been established. This means that collocations in definite descriptions either will arise only by accident (or by generate-and-test search) or by a secondary specificatioo that ensures the preference for semantics that can ultimately be realized using collocations.</Paragraph>
  </Section>
  <Section position="5" start_page="92" end_page="95" type="metho">
    <SectionTitle>
3 SPUD
</SectionTitle>
    <Paragraph position="0"> This section provides a brief overview of the representations and algorithms that Sentence Planning Using Description (SPUD) uses to address the properties of collocations discussed above. SPUD extends the general procedure for building referring expressions that is suggested by the planning paradigm (Appelt, 1985; Kronfeld, 1986). The procedure starts from a set of entities to describe and a set of intentions to achieve in describing them. It then applies operators that enrich the content of the description until all intentions are satisfied. As in realizations like (Dale and Haddock, 1991), we constrain the inference required to generate and evaluate alternatives by limiting the kinds of intentions considered. However, whereas the planning procedures on which we base our system are used only for noun phrases, we apply this procedure to the sentence as a whole using a rich semantic representation; further, although these procedures typically construct an abstract semantic representation, we treat operators as entries with syntactic, semantic and pragmatic properties. The lexicalized tree adjoining grammar (LTAG) formalism provides an abstraction of the combinatorial properties of words. The resulting system offers a number of advantages. By incorporating content into descriptions of a variety of entities until the addressee can fill in the details, this procedure results in short, natural and unambiguous sentences. Moreover, by evaluating and selecting alternatives on the basis of their pragmatic, semantic and syntactic contribution to the sentence as a whole, the procedure uniformly handles a variety of interactions inside a sentence, including collocations.</Paragraph>
    <Section position="1" start_page="92" end_page="93" type="sub_section">
      <SectionTitle>
3.1 Linguistic Specifications
</SectionTitle>
      <Paragraph position="0"> This algorithm requires a declarative specification of three kinds of information: first, what operators are available and how they may combine; second, how operators specify the content of a description; and third, how operators achieve pragmatic effects. We represent operators as elementary trees in an LTAG, and use TAG operations to combine them; we give the meaning of each tree as a formula in an ontologically promiscuous representation language; and, we model the praghmtics of operators by associating with each tree a set of discourse constraints describing when that operator can and should be used.</Paragraph>
      <Paragraph position="1"> TAG (Joshi et al., 1975) is a grammar formalism built around two operations that combine pairs of trees: SUBSTITUTION and ADJOINING. A TAG grammar consists of a finite set of ELEMEN-TARY trees, which can be combined by these operations to produce derived trees recognized by the grammar. In substitution, the root of the first tree is identified with a leaf of the second tree, called the substitution site (.L). Adjoining is a more complicated splicing operation, where the first tree replaces the subtree of the second tree rooted at a node called the adjunction site; that subtree is then substituted back into the first tree at a distinguished leaf called the FOOT node (,).</Paragraph>
      <Paragraph position="2"> Elementary trees without foot nodes are called INITIAL trees and can only substitute; trees with foot nodes are called AUXILIARY trees, and must adjoin. TAG elementary trees abstract the combinatorial properties of words in a linguistically appealing way. Figure 1 (a) shows an initial tree representing the book. Figure 1 (b) shows an auxiliary tree representing the modifier syntax, which could adjoin into the tree for the book to give the syntax book. All predicate-argument structures are localized within a single elementary tree, even in long-distance relationships. Figure l(c) shows the topicalized tree anchored by have; both of its arguments are substitution sites.</Paragraph>
      <Paragraph position="3"> Our grammar incorporates two additional principles. First, the grammar is LEXICALIZED (Schabes, 1990): each elementary structure in the grammar contains at least one lexical item.</Paragraph>
      <Paragraph position="4"> Second, our trees include FEATURES, following (Vijay-Shanker, 1987).</Paragraph>
      <Paragraph position="5"> We specify the semantics of trees by adapting two principles of computational semantics to the LTAG formalism. First, as originally advocated by Hobbs (1985), we adopt an ONTOLOGICALLY PROMISCUOUS representation that includes a wide variety of types of entities. In particular, abstract entities are introduced to represent the SCOPES of OPERATORS. A predicate is interpreted as if inside a scope when the predicate takes the corresponding abstract entity as an argument. For this paper, we need EVENTUALITIES as abstract representations of spatiotemporal scope and INFORMATION STATES to abstract the scope of modal operators like possibility and belief. Nodes are labeled as supplying information about a particular entity or  Nbbout : &lt;1&gt; L ?,X\] . I S SyrinX) ~ NPJ. \[about: &lt;2&gt; I,?,H-ee\] S \[about: &lt;1~\] N\[about. , , N* \[about:&lt;lq NP.L \[about: ?,?,.-er\] VP\[about : &lt;1~ syntax V \[ \] NP \[about : &lt;2~ concerns(l:INFO, S:STATE, I I X:IND, syntax:INb) /have./ t \[always applicable\] have(l:lNFO, H:STATE, H-er:IND, H-ee:INg)  (b) (in-poset(H-ee), in-op(have(I, H, H-er, H-ee))) (c)  collection of entities (this is inspired by a similar hypothesis in (Jackendoff, 1990)). To guarantee a coherent meaning for a derived structure, a node about x can only substitute or adjoin into another node about x. Here, we simply use an additional feature on the node to capture this. Figure 1 also shows the semantics and about labels for each tree; ? indicates unspecified about values.</Paragraph>
      <Paragraph position="6"> To package information appropriately requires sensitivity to the knowledge of the hearer and the state of the discourse. Different constructions make different assumptions about the status of entities and propositions. We model these differences by including in each tree a specification of the contextual conditions under which use of the tree is pragmatically licensed. Our conditions derive from linguistic analysis, particularly (Gundel et al., 1993; Ward, 1985; Ward and Prince, 1991; Prince, 1993; Birner, 1992).</Paragraph>
      <Paragraph position="7"> The status of entities and propositions in discourse varies along at least four dimensions that are relevant to these specifications. First, entities differ in NEWNESS (Prince, 1981). At any point, an entity is either new or old to the HEARER, according to whether or not the hearer has at least implicit knowledge of the existence of the entity. Analogously, an entity is either new or old to the DISCOURSE, according to whether the discourse contains an earlier reference to it. Second, entities differ in SALIENCE (Grosz and Sidner, 1986; Grosz et al., 1995). At any point, salience assigns each entity a position in a partial order that indicates how accessible it is for reference in the current context. Third, entities are related by material PARTIALLY-ORDERED SET (POSET) RELATIONS to other entities in the context (Hirschberg, 1985). These relations include part and whole, subset and superset, and membership in a common class; a number of constructions depend on poset relations to signal their connection with context. Finally, the discourse may distinguish some OPEN PROPOSITIONS, propositions containing free variables, as being under discussion (Halliday, 1967; Prince, 1986). This privileges subsequent information that provides true instantiations for the variables in a salient open proposition. We assume that information of these four kinds is available in a model of the current discourse state, and that the applicability conditions of constructions can freely make reference to this information. The pragmatic specification for the book, syntax, and topicalized have appear under the semantics for each tree in figure 1.</Paragraph>
      <Paragraph position="8"> Our discourse model contains information on the shared knowledge of the speaker and hearer, private knowledge of the speaker, and a specification of entities and their discourse status. In the library domain, shared knowledge includes such things as rules about how to check out books, while speaker knowledge includes such information as the status of books in the library. The discourse model can also include general properties that describe the conversational situation as a whole; for example, it might specify the formality of the register in which the communication is being conducted.</Paragraph>
    </Section>
    <Section position="2" start_page="93" end_page="94" type="sub_section">
      <SectionTitle>
3.2 The algorithm
</SectionTitle>
      <Paragraph position="0"> Our system takes two types of goals. First, goals of the form identify x as cat instruct the algorithm to construct a description of entity x using the syntactic category cat. If x is uniquely identifiable, then this goal is only satisfied when the overall content planned so far distinguishes x for the hearer. Ifx is hearer new, this goal is satisfied by including any constituent of type cat. Sec- null ond, goals of the form communicate p instruct the algorithm to include the proposition p. This goal is satisfied as.long as the overall content EN-TAILS p given the shared knowledge of speaker and hearer.</Paragraph>
      <Paragraph position="1"> In each iteration, our algorithm must determine the appropriate elementary tree to incorporate into the current description. It performs this task in two steps to take advantage of the regular associations between semantics and trees in the lexicon. Lexical entries pair a semantic constraint with a FAMILY of TREES that describe the combinatory possibilities for realizing the semantics. For example, book is stored with a tree family that includes a book and the book. We have chosen to include the determiners in the basic NP trees because of their importance for the semantics and pragmatics of the NP. Similarly, there are different initial trees for each clause type anchored by a particular verb. Trees in the tree family are shared among all lexical items that share a particular structure. This allows us to specify the pragmatic constraints associated with the tree type once and for all, regardless of which verb selects it. Moreover, we can determine which tree to use by looking at each tree ONCE, even when the same tree is associated with multiple lexical items.</Paragraph>
      <Paragraph position="2"> Hence, the first step is to identify applicable lexical entries: these items must correctly describe some entity; they must anchor trees that can substitute or adjoin into a node that describes the entity; and they must contribute toward satisfying current goals. (We describe more precisely how this contribution is evaluated in section 4.1 .) Then, the second step identifies which of the associated trees are applicable, by testing their pragmatic conditions against the current representation of discourse. We combine possible lexical items and possible trees, to give an evaluation of all applicable options. The algorithm identifies the entries that most contribute to current goals, and from these, selects the entry with the most specific semantic and pragmatic licensing conditions. This means that the algorithm generates the most marked licensed form for the particular context.</Paragraph>
      <Paragraph position="3"> The entry is then substituted or adjoined into the tree at the appropriate node. The meaning of the derived tree is simply the CONJUNCTION of the meanings of the elementary trees used to derive it. The entry may specify additional goals, because it describes one entity in terms of a new one. These new goals are added to the current goals, and then the algorithm repeats.</Paragraph>
    </Section>
    <Section position="3" start_page="94" end_page="95" type="sub_section">
      <SectionTitle>
3.3 Discussion
</SectionTitle>
      <Paragraph position="0"> The strength of the present work is that it captures a number of phenomena discussed elsewhere separately, and does so within the unified framework of description. In particular, we treat many types of content as contributing to expressions that refer to semantic objects. The tenses of sentences in discourse refer to times in much the same way pronouns and full NPs refer to individuals (Partee, 1973; Partee, 1984). The modality of sentences may refer to a salient possibility (Roberts, 1986) or provide the content of a salient psychological state (Wiebe, 1994). The rhetorical connection between a sentence and surrounding discourse should also be described with adjuncts (Huang, 1994). Adjuncts giving details about an event should be included only after reasoning that these adjuncts are in fact necessary in context (McDonald, 1992).</Paragraph>
      <Paragraph position="1"> With its incremental choices and its emphasis on the consequences of functional choices in the grammar, our algorithm resembles the networks of systemic grammar (Mathiessen, 1983; Yang et al., 1991). However, unlike systemic networks, our system derives its functional choices dynamically using a simple declarative specification of function that correlates well with recent linguistic work. Further, like many sentence planners, we assume that there is a flexible association between the content input to a sentence planner and the meaning that comes out. Other researchers (Nicolov et al., 1995; Rubinoff, 1992) have assumed that this flexibility comes from a mismatch between input content and grammatical options. In our system, such differences arise from the referential requirements and inferential opportunities that are encountered.</Paragraph>
      <Paragraph position="2"> Previous authors (McDonald and Pustejovsky, 1985; Joshi, 1987) have noted that TAG has many advantages for generation as a syntactic formalism, because of its localization of argument structure. These aspects of TAGs are crucial for us.</Paragraph>
      <Paragraph position="3"> Lexicalization allows us to easily specify local semantic and pragmatic constraints imposed by the lexical item in a particular syntactic frame.</Paragraph>
      <Paragraph position="4"> Various efforts at using TAG for generation (Mc-Donald and Pustejovsky, 1985; Joshi, 1987; Yang et al., 1991; Nicolov et al., 1995; Wahlster et al.,  1991) enjoy many of these advantages. Furthermore, (Shieber et al., 1990; Shieber and Schabes, 1991; Prevost and Steedman, 1993; Hoffman, 1994) exploit similar benefits of lexicalization and localization. What sets SPUD apart is its simultaneous construction of syntax and semantics, and the tripartite, lexicalized, declarative gram null matical specifications for constructions it uses.</Paragraph>
      <Paragraph position="5"> (Shieber et al., 1990; Shieber and Schabes, 1991 ) construct a simult .~eous derivation of syntax and semantics--but they do not construct the semantics: it is an input to their system. Moreover, they do not represent any pragmatic inforrnatiGn.</Paragraph>
      <Paragraph position="6"> (Prevost and Steedman, 1993; Hoffman, 1994) do represent the division of sentences into theme and rheme, but because they do not model the pragmatics of particular constructions, they plan descriptions in a separate step.</Paragraph>
    </Section>
  </Section>
  <Section position="6" start_page="95" end_page="96" type="metho">
    <SectionTitle>
4 Conventional combination in SPUD
</SectionTitle>
    <Paragraph position="0"> Because LTAG can associate multiple iexical items to a single tree, it is straightforward to list frozen idioms, like call number, in the lexicon (Abeille and Schabes, 1989). These specifications can include idiosyncratic semantic and pragmatic information; grammatical processes like tense marking apply normally.</Paragraph>
    <Paragraph position="1"> In this section, we describe how SPUD can be made to use words in other conventional combinations. Our proposal involves three steps. First, as in (Reiter and Dale, 1992), we stipulate that some attributes of entities are more important than others, and that some words more naturally describe those attributes. Second, in keeping with ontological promiscuity (Hobbs, 1985), we represent the importance of attributes by the salience of events and states in the discourse model--these states and events now have the same status in the discourse model as any other entities. Finally, we extend SPUD's evaluation of alternatives, so that it describes the most salient entities possible, and uses basic-level terms wherever possible. By associating entities not just with salient attributes but also with salient actions and salient figurations, we capture collocations, semantic collocations and idiomatic compositionality using a uniform mechanism.</Paragraph>
    <Section position="1" start_page="95" end_page="96" type="sub_section">
      <SectionTitle>
4.1 Collocations proper
</SectionTitle>
      <Paragraph position="0"> Although primarily concerned with the interpretation of Gricean maxims, the work of (Reiter and Dale, 1992; Dale and Reiter, 1995) underlines the conventionality of description. Based on a review of psychological experimentation and their own study of referring expressions in task-oriented dialogue, they argue that some referring expressions can be constructed simply by selecting properties from a prioritized list of attributes until the entity is distinguished. To further conventionalize descriptions, they privilege the selection of properties that provide basic-level characterizations of the entity (Rosch, 1978; Reiter, 199l). Because any property is considered for only one attribute, this algorithm offers a linear speedup over the greedy strategy used in (Dale and Haddock, 1991) and described above for SPUD, which considers every property at every stage. However, here we focus on how incorporating similar ideas into SPUD gives a general framework for specifying conventional uses of words, and remain neutral about achieving similar speedups.</Paragraph>
      <Paragraph position="1"> Reiter and Dale suggest that the prioritized list of attributes their algorithm uses is domaindependent. In fact, we find that these lists are both domain and object-dependent. Obviously the attributes by which we describe abstractions like events and states--typically time, location, and manner or quality--are quite distinct from the natural attributes by which physical objects are distinguished. However, in the library, widely different attributes can be appropriate even for physical objects of various types. Books can be described by author, by physical characteristics, or by content (e.g. Chomsky ~ book; the yellow book, a math book). Periodicals, meanwhile, are best described by date of issue (e.g. the May issue of Language). Parts of the library, as we shall see below, are best distinguished by the special services they provide (e.g. the reference desk).</Paragraph>
      <Paragraph position="2"> SPUD's ontologically promiscuous discourse model offers a natural dimension to represent these distinctions. Since each property of an object is associated with an eventuality argument, we can assign a level of salience for that eventuality. We can use this ranking to indicate the conventional importance of the eventuality in distinguishing the object. In other words, if we know p(e, x), and it is natural to describe x in terms of p, e will be salient. For example, since periodicals are easily identified by their date of issue, we should make this state salient. Note then that salience is determined for explicitly mentioned and inferable entities and depends not only on recency of mention but also on facts about the conversational situation and real-world relationships between objects.</Paragraph>
      <Paragraph position="3"> Reiter and Dale also point out that which characterizations are basic-level must be adjusted to reflect the expertise of the addressee; however, we shall sidestep this issue here by assuming that certain lexical items are simply listed as basic-level terms.</Paragraph>
      <Paragraph position="4"> By itself, these additions are not enough: SPUD must also take salience and basic-level semantics into account in the evaluation of its alternatives. That is: other things being equal, SPUD should choose to incorporate at each stage the  syntactic-semantic-pragmatic unit which refers to maximally salient entities; and, other things being equal, SPUD should incorporate a basic-level predicate. Integrating Reiter and Dale's prioritization of these considerations with SPUD's other considerations leads to the following ranking of criteria for comparison:</Paragraph>
    </Section>
  </Section>
  <Section position="7" start_page="96" end_page="98" type="metho">
    <SectionTitle>
(7) RULES OUT A DISTRACTOR OR ENTAILS
NEEDED INFORMATION &gt; SALIENCE OF
ENTITIES MENTIONED &gt; NUMBER OF
DISTRACTORS RULED OUT &gt; NUMBER OF
INFORMATIONAL GOALS ACHIEVED &gt;
BASIC-LEVEL TERM &gt; SPECIFICITY OF
LICENSING CONDITIONS
</SectionTitle>
    <Paragraph position="0"> With the right linguistic specification, this is all the machinery SPUD needs to generate conventionalized forms. To see how we can generate ordinary collocations, consider describing parts of a library. Descriptions of these places are typically collocations: e.g. copy area, reference desk, interlibrary loan office. The names can be abbreviated in context, they can be interpreted compositionally, but substituting synonyms generally sounds odd. Nevertheless, these descriptions share features, in that one always describes its type, sometimes the service it provides, and most rarely its location. This leads to the following axiomatization of the salience of states:</Paragraph>
    <Paragraph position="2"> The first argument of each predicate is the information state in which the various predications hold; the second argument is the eventuality which witnesses the application of the predicate; &gt;s indicates the salience ranking of the states.</Paragraph>
    <Paragraph position="3"> Thus, (8) considers a case where there is a part Part of a library Lib: suppose $3 witnesses that Part has some type Type; $4, that Part provides service Service; and $5, that Part has location Loc. Then, $4 is more salient than $5, and $3 is more salient than both. We must specify not only the salience of different states for the same copier, but also the salience of corresponding states for different copiers. Another axiom, similar to (8), ensures that states that specify a given attribute are equally salient across copiers when the copiers involved are equally salient.</Paragraph>
    <Paragraph position="4"> The vocabulary chosen, meanwhile, reflects conventional names for the structures and services of the library. Semantic declarations such as the following represent this: (9) area (I, S, A) : BASIC has-type(I, S, A, area) That is, area uses the specified semantics to provide a basic-level description of A in terms of state 9 and information I. Note that SPUD always chooses a maximally specific licensed form out of equally good alternatives. Thus, we can have any number of basic-level terms to describe an object, and the appropriate one will be selected on the basis of its specificity. For example, even if both room and area are basic, a room will be still be described using room, because all rooms are areas but not all areas are rooms.</Paragraph>
    <Paragraph position="5"> Together, these assumptions suffice to generate collocations for library parts. For example, suppose SPUD has the goal of describing the part of the library where copying takes place, loca-tion e30. SPUD first selects the NP the area, eliminating alternatives like the room, the desk, the stack, because they do not truthfully describe e30. However, since many other parts of the library are also areas, the current description does not rule out all possible distractors, and SPUD further elaborates the description. The modifiers copy and service are both applicable to e30, but copy eliminates all distractors while service does not, so the former is selected, yielding the final NP the copy area.</Paragraph>
    <Section position="1" start_page="96" end_page="97" type="sub_section">
      <SectionTitle>
4.2 Semantic collocations
</SectionTitle>
      <Paragraph position="0"> To handle semantic collocations now requires only a representation of how certain lexical items depend on hidden parameters for actions and events. For example, consider the lexical item fast: it constrains the typical rate of some action performed by or with the entity it describes. Thus, it has a meaning like this:</Paragraph>
      <Paragraph position="2"> Corresponding to the qualia structure of GLT, we have axioms describing what actions are associated with objects and how salient they are. For a photocopier, this might be specified this way: (1 I) photocopier(I, S, X) 3 (participant ( I, S 1 (X), X, copy-action) A participant(I, S2(X), X, repair-action) A participant(I, S3(X), X, fill-paper-action)A SI(X) &gt;s S2(X) &gt;s S3(X)) That is, typically, with copiers, you not only make copies, but also fill them with paper, and (sadly, all too often), have them repaired; however, copying is the most salient thing to do with them. Note that while this axiom is expressed at the same level of  generality as GLT's qualia structures, this rule is part of world knowledge and applies to all things that are photocopiers, not to all occasions where things are described as photocopiers.</Paragraph>
      <Paragraph position="3"> To see how SPUD uses these specifications, let us say that we have a copier, c42, which is the sole fast copier (at making copies) in the library. After planning a refemng expression the copier, SPUD has the goal of distinguishing c42 from the other copiers. The KB entails the fact fast(i,s,c42,copy-action), which allows us to incorporate the lexical item fast into the description. SPUD then evaluates the distractor set; since copy-action is a new reference, SPUD checks whether any distractor is also fast at an action which is at least as salient as copy-action. None are, because copy-action is the most salient action of copiers. Since the expression, the fast copier, now refers uniquely both to c42 and to copy-action, the referring expression is adequate. The need to rule out distractor actions can cause information to be added to an expression. To describe another copier, c43, which is the fastest copier to fill with paper, SPUD would describe not only its rate but also the relevant action in order to distinguish it from c42, i.e.</Paragraph>
      <Paragraph position="4"> the fast copier to fill. Also, note SPUD can use this same meaning of fast and the same reasoning process even when fast does not modify a noun. (For example, in a slightly different context it could describe the state s with this sentence: The copier is fasr)</Paragraph>
    </Section>
    <Section position="2" start_page="97" end_page="98" type="sub_section">
      <SectionTitle>
4.3 Idiomatic composition
</SectionTitle>
      <Paragraph position="0"> As (Nunberg et al., 1994) emphasize, idiomatic composition typically involves some distinctive figurative or metaphorical view of the objects being described. Accordingly, to specify idiomatic composition, we adopt a representation of such views from (Ballim et al., 1991). They outline a model of reasoning in which facts are partitioned into sets called ENVIRONMENTS. Environments can collect information about particular topics, or, when nested, can represent the beliefs of particular agents. Moreover, they suggest that non-literal language can also be represented using a nested environment, whose contents are determined by treating topic-environments as competing sources of information analogous to different agents' views. We believe reasoning algorithms like those presented in (Ballim et al., 1991) should be an important part of any natural language generation system which aims at idiomatic language; however, for the present, the key feature of this account is just its principled use of multiple informatiou-states, in which different facts hold.</Paragraph>
      <Paragraph position="1"> We combine this representation with two assumptions about how information states are represented in the grammar. We assume that information states are recovered from the context just like other parameters of interpretation like states and actions. However, we use trees that in some cases impose coreference requirements between the information states in which different constituents are interpreted. For the examples we have considered, what seems right is to coindex the information states of modifiers and their heads, and to coindex the information state of a verb with all its arguments except the subject.</Paragraph>
      <Paragraph position="2"> (The trees of figure 1 respect this generalization.) Consider the example from section 2: the combined convention strings = influence, pull = exert privately. The opportunity to use the expression arises in any information state k where: (12) influence(k, Sl, C, X, F) A subverts(k, $2, C, bureaucracy) A exert(k, E, X, C) A private(k, $3, E) We can represent the idiom semantically using a rule that introduces the associated stock figuration, that bureaucrats are puppets whose behavior is governed by such influence: bp(k, C).</Paragraph>
      <Paragraph position="3"> (13) strings(bp(k, C), S4(k, C), C) A pull(bp(k, C), E, X, C) Now we just use the ordinary meanings of pull and strings to describe this situation.</Paragraph>
      <Paragraph position="4"> To constrain the situations in which this is an appropriate thing to say, we need to determine the circumstances in which bp(k, C) is as salient as k. (One might claim that the ready salience of the information state--naturally, different across languages--is what makes idioms different from metaphors.) Although such a specification is clearly open-ended, we approximate the full set of constraints in terms of two parameters of the discourse context: a reasonable degree of intimacy between speaker and hearer and an informal register of conversation.</Paragraph>
      <Paragraph position="5"> Consider how the noun phrase the strings she pulled is generated to describe some exerted influence c. Under appropriate discourse conditions, SPUD can choose to describe c in terms of the information state bp(k,c) and the lexical item strings. To rule out c's additional distractors, the object relative clause anchored by pulled is chosen; the informational coindexation between the foot N node and the verb in an object relative clause ensures that exerted does not apply-because c is NOT the object of an exerting event according to information bp(k,c). Finally, the  agent of the pulling is described with she.</Paragraph>
    </Section>
  </Section>
class="xml-element"></Paper>