Stone, M., Doran, C.: Sentence planning as description using tree-adjoining grammar. In: Proceedings of the 35th Annual Meeting of the ACL. (1997) 198--205  Home/Search   Document Details and Download   Summary   Related Articles   Check

....is a binary tree with leaves labeled by all the elementary speech acts from the input text plan, and with its interior nodes labeled with clause combining operations. The sp tree is inspired by Lavoie and Rambow [22] The representations used by Danlos, Gardent and Webber, and Stone and Doran [9, 13, 42] are similar, but do not (always) explicitly represent the clause combining operations as labeled nodes. In our representation of the sp tree, each node is also associated with a DSyntS: the leaves (which correspond to elementary speech acts from the input text plan) are linked to a canonical ....

Matthew Stone and Christine Doran. Sentence planning as description using tree adjoining grammar. In Proceedings of the 35th Annual Meeting of the Association of Computational Linguistics, ACL/EACL 97, pages 198-205, Madrid, Spain, 1997.

....gesturing, Rea allows herself to be interrupted, and then takes the turn again when she is able. She is able to initiate conversational repair when she misunderstands what the user says, and can generate combined voice and gestural output. An incremental natural language generation engine based on (Stone Doran, 1997), and extended to synthesize redundant and complementary conversational hand gestures, generates Rea s responses. Figure 6 shows Figure 4: User interacting with Rea Cassell Bickmore Negotiated Collusion 14the architecture for the Rea system. REA is an acronym for Real Estate Agent , and ....

Stone, M., & Doran, C. (1997). Sentence Planning as Description Using Tree-Adjoining Grammar, ACL pp. 198--205, Madrid, Spain: MIT Press.

....and the amount of knowledge they needed was too vast, to support their use in anything other than a toy system or PhD thesis. This situation is now changing: in Natural Language generation, fast automated reasoners are being used in integrated systems for micro planning (Stone, 1998; Stone, 2000; Stone and Doran, 1997; Stone and Webber, 1998; Stone et al. 2001; Gardent and Striegnitz, 2001) while in Natural Language understanding the concern of this paper fast automated reasoners are being employed over an ever expanding range of problems. In this paper, we focus on the role inference can play in ....

Stone, M. and C. Doran: 1997, `Sentence Planning as Description Using Tree Adjoining Grammar'. In: Proceedings of the 35 th Annual Meeting of the Association for Computational Linguistics (ACL97/EACL97). Madrid, Spain, pp. 198-205.

.... architecture is needed which interleaves rather than pipeline contextual reasoning (e.g. the computation of a uniquely identifying set of properties) and sentence realisation (e.g. the surface realisation of a de nite NP) Speci cally, they show that the SPUD architecture presented in Stone and Doran s (1997) appropriately captures the type of textual economy illustrated by the above examples. In this paper, we follow Stone and Webber (1998) and adopt a SPUDlike architecture to generate indirect anaphors. This architecture consists of three main components: A Lexicalised Tree Adjoining Grammar ....

....NP Det the N# N rabbit = NP Det the N rabbit substitute adjunction: NP Det the N rabbit N Adj white N = NP Det the N Adj white N rabbit adjoin Figure 1: Examples of substitution and adjunction. The lexicon we use for generation follows the guidelines set in Stone and Doran s (1997) SPUD generation system. In particular, it has the following properties: The trees are associated with semantic and pragmatic information. The semantic representation language is a at semantics (Hobbs 1985; Copestake et al. 1999) The semantics of a complex tree is just the union of the ....

Stone, M. and C. Doran (1997). Sentence Planning as Description Using Tree Adjoining Grammar. In Proceedings of ACL, pp. 198-205.

....operators on predicates. Such a flat representation is also motivated in the context of generation, where one wants to start with a representation of the input which makes the minimal commitment to structure. Details on associating a flat semantics with a derived TAG tree can also be found in (Stone Doran, 1997) and (Stone et al. 2001) In (7) below, we roughly illustrate the semantic formula associated with the example discourse in (4) We refer to the semantics of the connective trees by the names of the connectives, and use to represent the semantics associated with the auxiliary tree associated ....

Stone, Matthew & Christine Doran (1997). Sentence Planning as Description using Tree Adjoining Grammar. In Proc. of ACL, pp. 198--205.

....model in a generation system. As underlying formalism of our integrated generation model we have chosen Schema TAGs with Unification (SU TAGs) 1 because TAGs provide the necessary complexity to express any kind of concept in the what to say and how to say component (cf. e.g. (Stone Doran, 1997), Webber Joshi, 1998) Becker et al. 1998) Nicolov, 1998) Schema TAGs are especially 1 In a schematic elementary tree, a regular expression (RX) is annotated at each inner node of an elementary tree. This means, that the elementary schemata enumerate a possibly infinite set of ....

....transformation algorithms for the individual knowledge bases of a generation system must be provided. Any TAG can automatically be transformed into a SU TAG (Harbusch et al. 1998) This is already done for the syntactic knowledge base XTAG (Doran et al. 1994) The knowledge sources of SPUD (Stone Doran, 1997), those of anchored L TAGs (cf. Webber Joshi, 1998) as well as the TAG transformed from an HPSG (Becker et al. 1998) will be rewritten as SU TAG next. Doing so, the generation system is extended towards a generation workbench which provides libraries with knowledge sources from which the ....

STONE M. & DORAN C. (1997). Sentence planning as description using Tree Adjoining Grammar. In Procs. of the 35th ACL and 8th EACL, Madrid, Spain.

....true, at some level, for discourse syntax and semantics. Here we demonstrate our initial e ort to extend a lexicalized grammar (LTAG) to discourse; use the same compositional semantics on syntactic structure that is used in lexicalized grammars (Steedman, 1996; Joshi and VijayShanker, 1999; Stone and Doran, 1997; Stone and Webber, 1998) extend to discourse connectives the idea that the meaning of a lexical element can involve an anaphoric link to the previous discourse; exploit similar inference mechanisms for defeasible aspects of both sentence level and discourse semantics. 1 The original paper that ....

Stone, M. and C. Doran: 1997, `Sentence Planning as Description Using Tree Adjoining Grammar'. In: Proceedings of the 35 th Annual Meeting of the Association for Computational Linguistics (ACL97/EACL97). Madrid, Spain, pp. 198-205.

....gesturing, Rea allows herself to be interrupted, and then takes the turn again when she is able. She is able to initiate conversational repair when she misunderstands what the user says, and can generate combined voice and gestural output. An incremental natural language generation engine based on [28], and extended to synthesize redundant and complementary conversational hand gestures, generates Rea s responses. REA is an acronym for Real Estate Agent , and within this domain we are currently focused on modeling the initial interview with a prospective buyer. Real estate sales was selected ....

Stone, M. and Doran, C., "Sentence Planning as Description Using Tree-Adjoining Grammar," presented at ACL, 1997.

.... expression are often used interchangeably though they mean different things: not all nominal expressions or noun phrases actually refer to an identifiable entity or set, and some researchers have generalised the notion of reference so that for instance sentences may be said to refer (see e.g. Stone and Doran 1998; see also van Deemter and Kibble (to appear) for more general discussion) In fact this report is mostly concerned with referential uses of nominal expressions, particularly 4 anaphoric uses. That is, we make the simplifying assumption (common in the NLG literature) that all NPs in a generated ....

M Stone and C Doran 1998. Sentence Planning as Description Using Tree Adjoining Grammar. Proceedings of ACL-EACL '97.

....what particular agents know about actions in the domain are discussed. These rules for reasoning about actions and agents both contribute to generation of effective instructions. 1. 3 Generating Effective Instructions A Natural Language generator called SPUD (Sentence Planning Using Descriptions) [Stone and Doran, 1997; Stone and Webber, 1998; Stone, 1998] developed at the University of Pennsylvania, provides the necessary components for generating effective instructions. SPUD forms descriptions of actions (as well as events, states, and objects) by choosing lexical items from its Lexicalized Tree Adjoining ....

....instructions involving expressions of action termination. 49 Chapter 4 Representing Objects, Actions, and Agent Expertise The Natural Language generator SPUD (Sentence Planning Using Descriptions) Stone, 1998] reasons about Natural Language Generation (NLG) tasks using a logic theoremprover [Stone, 1997], presenting the opportunity to develop domain knowledge which is independently motivated. Implementing a domain model in SPUD is much like writing a logic program since it uses modal first order logic, a combination of first order logic and modal logic (see Section 2.3) Therefore, SPUD can be ....

[Article contains additional citation context not shown here]

M. Stone and C. Doran. Sentence planning as description using Tree-Adjoining Grammar. In Proceedings of ACL/EACL'97, pages 198--205, 1997.

....domain propositions that encode specified kinds of information about a specified object. The communicative action generated must also fit the context specified by the discourse model, to the best extent possible. We use the SPUD generator ( Sentence Planning UsinDescription ) introduced in Stone and Doran (1997) to carry out this generation task. Figure X.5 shows the structure of the simultaneous speech and gesture generation process in the Generation Module. An utterance generation process starts when the Decision Module sends out a generation speech act. The generation speech act is usually in the ....

Stone, M., and C. Doran. 1997. Sentence planning as description using tree adjoining grammar. In Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics and 8th Conference of the European Chapter of the Association for Computational Linguistics, 198--205, Madrid, Spain.

No context found.

M. Stone and C. Doran. Sentence planning as description using treeadjoining grammar. In ACL, pages 198--205, 1997.

....invite the hearer to conclude about those individuals how can this sentence be modified or extended can the generator recognize and exploit an opportunity for textual economy. These representational and reasoning requirements are met in the SPUD system for sentence planning and realization [26, 27]. SPUD draws on earlier work by Appelt [1] in building sentences Using planning techniques. sPUD plans the syntax and semantics of a sentence by incorporating lexico grammatical entries into a partial sentence one by one and incrementally assessing the answers to the questions given above. In this ....

....lexico grammatical entries into a partial sentence one by one and incrementally assessing the answers to the questions given above. In this paper, we describe the intermediate representations that allow sPUD to do so, since these representations have been glossed over in earlier presentations [26, 27]. Reasoning in SPUD is performed using a fast modal theorem prover [24, 25] to keep track both of what the sentence entails and what the sentence requires in context. By reasoning about the predicated relationships withinclauses and the informational relationships [16] between clauses, sPUD is ....

[Article contains additional citation context not shown here]

Matthew Stone and Christine Doran. Sentence planning as description using tree-adjoining grammar. In Proceedings of ACL, pages 198-205, 1997.

....judge that the hearer can use shared information, the utterance, the grammar, and the attentional and intentional state of the discourse, to reconstruct this interpretation. This is the formulation of the language production problem that I and my colleagues arrived at in the SPUD generation system [Stone and Doran, 1997, Stone and Webber, 1998, Stone et al. 2001] SPUD can generate concise, contextually appropriate utterances, including both speech and concurrent nonverbal behavior, by applying a simple, uniform and efficient decision making strategy. This strategy exploits the lexicalized tree adjoining ....

Stone, M. and Doran, C. (1997). Sentence planning as description using treeadjoining grammar. In Proceedings of ACL, pages 198--205.

No context found.

Matthew Stone and Christine Doran. 1997. Sentence planning as description using tree adjoining grammar. In Proc. 35 th Annual Meeting of the Association for Computational Linguistics, pages 198--205, Madrid, Spain. Morgan Kaufmann.

No context found.

Matthew Stone and Christine Doran. Sentence planning as description using tree-adjoining grammar. In Proceedings of ACL, pages 198--205, 1997.

....among them: 3) Remove the bolt with the wrench from the tool box. In (3) one compact and natural sentence describes the goal of an action (removal) its method (wrench) and the knowledge required to accomplish it (the wrench s location) The sentence planning using description (spud) system [27, 28] I describe in section 3.3 also assigns inference about interpretation a central role in generation, albeit a more constrained role than Appelt. These arguments show the advantages that we can expect inference to o#er in ambitious dialogue systems. But what about the kinds of NLG systems that ....

....Sections 3.1 and 3.2 have shown how inferences about knowledge and action can be specified and derived. It remains to be seen how these conclusions might actually be accessed in sentence planning. In this section, I suggest one answer, inspired 1 by the spud system for sentence generation [27, 28]. spud adopts a view of sentence generation as goal directed activity, like [1, 4] before it. On this view, the task of the generator is to use the words and constructions of the language to design a message that fulfills a set of communicative intentions. spud works with two kinds of intentions ....

Matthew Stone and Christine Doran. Sentence planning as description using tree-adjoining grammar. In Proceedings of ACL, pages 198--205, 1997.

....descriptions by assessing how the interpretation of a verb in context contributes towards the hearer s identification of the intended action. 1. Introduction This paper originates in a project of tailoring a natural language generation system called SPUD, for sentence planning using description (Stone Doran, 1997), to generate instructions for action in a concrete domain. The desired behavior for the system is specified by a corpus of edited, naturally occurring action instructions whose form and content the system must mirror. The input to the system consists of three components: a representation of the ....

....it selects leads to the construction of generation resources that can account for such alternation. First, observe that the syntax and the syntax semantics interface put slide and position on an equal footing. We can settle on a syntactic tree for each verb that best fits the context as in (Stone Doran, 1997); we have designed these trees so that either choice can be fleshed out by further constituents into a satisfactory utterance. To choose one verb in construction over the other, we must look at the INTERPRETATION of the two entries. A key part of this interpretation is the way the hearer resolves ....

STONE M. & DORAN C. (1997). Sentence planning as description using tree-adjoining grammar. In ACL, p. 198--205.

No context found.

Stone, M., Doran, C.: Sentence planning as description using tree-adjoining grammar. In: Proceedings of the 35th Annual Meeting of the ACL. (1997) 198--205

No context found.

Stone, M. and Doran, C., Sentence Planning as Description Using TreeAdjoining Grammar. Proceedings of ACL 1997, pages 198--205.

No context found.

Stone, Matthew & Christine Doran (1997). Sentence Planning as Description using Tree Adjoining Grammar. In Proc. of ACL, pp. 198--205.

No context found.

Matthew Stone and Christine Doran. Sentence planning as description using tree adjoining grammar. In Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics, pages 198--205, Madrid, Spain, 1997. The Association for Computational Linguistics.

No context found.

Matthew Stone and Christine Doran. Sentence planning as description using tree adjoining grammar. In Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics, pages 198--205, Madrid, Spain, 1997. The Association for Computational Linguistics.

No context found.

Matthew Stone and Christine Doran. 1997. Sentence planning as description using tree adjoining grammar. In 35th Meeting of the Association for Computational Linguistics (ACL'97), pages 198--205, Madrid, Spain.

No context found.

Matthew Stone and Christine Doran. 1997. Sentence Planning as Description using Tree Adjoining Grammar. In Proceedings of ACL-EACL '97, Madrid, Spain.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC