| { | |
| "paper_id": "J82-3008", | |
| "header": { | |
| "generated_with": "S2ORC 1.0.0", | |
| "date_generated": "2023-01-19T03:07:14.488107Z" | |
| }, | |
| "title": "The FINITE STRING Newsletter Abstracts of Current Literature Planning Natural Language Utterances", | |
| "authors": [ | |
| { | |
| "first": "Douglas", | |
| "middle": [ | |
| "E" | |
| ], | |
| "last": "Appelt", | |
| "suffix": "", | |
| "affiliation": {}, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Marc", | |
| "middle": [], | |
| "last": "Luria", | |
| "suffix": "", | |
| "affiliation": {}, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Lawrence", | |
| "middle": [], | |
| "last": "Birnbaum", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "Yale Station Yale University New Haven", | |
| "location": { | |
| "postBox": "Box 2158", | |
| "postCode": "06520", | |
| "settlement": "Connecticut" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "James", | |
| "middle": [ | |
| "F" | |
| ], | |
| "last": "Allen", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "The University of Rochester", | |
| "location": { | |
| "postCode": "14627", | |
| "settlement": "Rochester", | |
| "region": "New York" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Alan", | |
| "middle": [ | |
| "M" | |
| ], | |
| "last": "Frisch", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "The University of Rochester", | |
| "location": { | |
| "postCode": "14627", | |
| "settlement": "Rochester", | |
| "region": "New York" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Diane", | |
| "middle": [ | |
| "J" | |
| ], | |
| "last": "Litman", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "The University of Rochester", | |
| "location": { | |
| "postCode": "14627", | |
| "settlement": "Rochester", | |
| "region": "New York" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "David", | |
| "middle": [ | |
| "D" | |
| ], | |
| "last": "Mcdonald", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "University of Massachusetts Amherst", | |
| "location": { | |
| "postCode": "01002", | |
| "settlement": "Massachusetts" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "E", | |
| "middle": [ | |
| "Jeffery" | |
| ], | |
| "last": "Conklin", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "University of Massachusetts Amherst", | |
| "location": { | |
| "postCode": "01002", | |
| "settlement": "Massachusetts" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Mark", | |
| "middle": [ | |
| "A" | |
| ], | |
| "last": "Jones", | |
| "suffix": "", | |
| "affiliation": {}, | |
| "email": "" | |
| }, | |
| { | |
| "first": "David", | |
| "middle": [ | |
| "S" | |
| ], | |
| "last": "Warren", | |
| "suffix": "", | |
| "affiliation": {}, | |
| "email": "" | |
| }, | |
| { | |
| "first": "David", | |
| "middle": [ | |
| "I" | |
| ], | |
| "last": "Keirsey", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "University of California", | |
| "location": { | |
| "postCode": "92717", | |
| "region": "California" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Laura", | |
| "middle": [], | |
| "last": "Silver", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "University of Pittsburgh Pittsburgh", | |
| "location": { | |
| "country": "Pennsylvania" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Lawrence", | |
| "middle": [ | |
| "J" | |
| ], | |
| "last": "Mazlack", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "University of Cincinnati Cincinnati", | |
| "location": { | |
| "postCode": "45221", | |
| "region": "Ohio" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Noemi", | |
| "middle": [ | |
| "M" | |
| ], | |
| "last": "Paz", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "University of Cincinnati Cincinnati", | |
| "location": { | |
| "postCode": "45221", | |
| "region": "Ohio" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Stuart", | |
| "middle": [], | |
| "last": "Shapiro", | |
| "suffix": "", | |
| "affiliation": {}, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Joao", | |
| "middle": [], | |
| "last": "Martins", | |
| "suffix": "", | |
| "affiliation": {}, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Donald", | |
| "middle": [], | |
| "last": "Mckay", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "Research and Development Activity Special Systems Division Federal and Special Systems Group", | |
| "institution": "", | |
| "location": { | |
| "postBox": "P.O. Box 517", | |
| "postCode": "19301", | |
| "settlement": "Paoli", | |
| "region": "Pennsylvania" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Anthony", | |
| "middle": [], | |
| "last": "Jameson", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "Research Unit for Information Science and Artificial Intelligence", | |
| "institution": "University of Hamburg", | |
| "location": { | |
| "addrLine": "Mittelweg 179", | |
| "postCode": "D-2000", | |
| "settlement": "Hamburg 13", | |
| "country": "WEST GERMANY" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Wolfgang", | |
| "middle": [], | |
| "last": "Wahlster", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "Research Unit for Information Science and Artificial Intelligence", | |
| "institution": "University of Hamburg", | |
| "location": { | |
| "addrLine": "Mittelweg 179", | |
| "postCode": "D-2000", | |
| "settlement": "Hamburg 13", | |
| "country": "WEST GERMANY" | |
| } | |
| }, | |
| "email": "" | |
| } | |
| ], | |
| "year": "", | |
| "venue": null, | |
| "identifiers": {}, | |
| "abstract": "This paper describes recent research on a naturallanguage-generation system that is based on planning. A system named KAMP is described that is capable of producing English sentences as part of a plan to enlist the cooperation of another agent in achieving a goal involving a change in the physical state of the world. The planner uses knowledge about the different subgoals to be achieved and linguistic rules about English to produce sentences that satisfy multiple goals through the realization of multiple illocutionary acts. Abstracts of Current Literature \\ is a progress report on the project, called ARGOT. It outlines the system and describes recent results as well as work in progress.", | |
| "pdf_parse": { | |
| "paper_id": "J82-3008", | |
| "_pdf_hash": "", | |
| "abstract": [ | |
| { | |
| "text": "This paper describes recent research on a naturallanguage-generation system that is based on planning. A system named KAMP is described that is capable of producing English sentences as part of a plan to enlist the cooperation of another agent in achieving a goal involving a change in the physical state of the world. The planner uses knowledge about the different subgoals to be achieved and linguistic rules about English to produce sentences that satisfy multiple goals through the realization of multiple illocutionary acts. Abstracts of Current Literature \\ is a progress report on the project, called ARGOT. It outlines the system and describes recent results as well as work in progress.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Abstract", | |
| "sec_num": null | |
| } | |
| ], | |
| "body_text": [ | |
| { | |
| "text": "paradigm and Montague Grammar (MG) formalism. Although these two approaches may seem to be strange bedfellows indeed with often noticeably different perspectives, we have observed many commonalities. We begin with a brief description of the problem view and ontology of each and then create a formulation of CD as logic. We then give \"conceptual\" MG translations for the words in an example sentence which we use in approximating a word-based parsing style. Finally, we make some suggestions regarding further extensions of logic to introduce higher level representations. Proc. 1982 AAAI Conf., Aug. 1982 \"Event shape diagrams\" are proposed as a representation for capturing the nuances of meaning of verbs that describe similar events. These diagrams represent timing, causal relationships between case roles, and typical value ranges for role fillers. Event shape diagrams are expressed in terms of primitive predicates and timing information that we believe could be computed by perceptual systems, and are intended to be a step toward the eventual connection of language systems to perceptual (vision, hearing, and touch) sensing systems. The diagrams are capable of representing modification of verbs by adverbs, can support judgments of the degree of plausibility of various interpretations of a sentence's meaning, and may be useful in figuring out the meaning of certain kinds of metaphors. Proc. 1982 AAAI Conf., Aug. 1982 This paper describes the principles of Right Association and Minimal Attachment and explains how the Sausage Machine and ATN describe these principles. It is then shown that these two models cannot explain these principles. It is then shown that a production system grammar can both describe these principles as well as suggest why they must be true. Proc. 1982 AAAI Conf., Aug. 1982 A representation and processing scheme for temporal (time-based) information is presented. Previous computer science approaches to temporal information processing are analyzed. Linguistic analysis of tense, aspect, and temporal adverbials provides motivation for an automated general temporal understanding system. A synthetic approach is proposed, combining possible-worlds branching time theory with inertia futures, elements of Montague Grammar, a fourvalued logic and the interval semantics time model. Key portions of the model are implemented and demonstrated in a PASCAL program.", | |
| "cite_spans": [ | |
| { | |
| "start": 573, | |
| "end": 583, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 584, | |
| "end": 605, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1400, | |
| "end": 1410, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1411, | |
| "end": 1432, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1784, | |
| "end": 1794, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1795, | |
| "end": 1816, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Providence, Rhode Island 02912 Proc. 1982 AAA/ Conf., Aug. 1982 The tasks of disambiguating words and determining case are similar and can usefully be combined. We present two cooperating mechanisms that each work on both tasks: MARKER PASSING finds connections between concepts in a system of frames, and POLAR-OID WORDS provide a protocol for negotiation between ambiguous words and cases. Examples of each in action are given. The cooperating mechanisms allow linguistic and world knowledge to be unified, frequently eliminate the need to use inference in disambiguation, and provide a usefully constrained model of disambiguation. Proc. 1982 AAAI Conf., Aug. 1982 UC (UNIX Consultant) is an intelligent natural language interface that allows naive users to communicate with the UNIX operating system in ordinary English. UC is currently capable of handling simple dialogues, including ones which require awareness of the context. UC is being extended to handle requests requiring more complex reasoning to formulate an intelligent response. Proc. 1982 AAAI Conf., Aug. 1982 Efficient syllabic hypothesization in continuous speech has been so far an unsolved problem. A novel solution based on the extraction of acoustic cues is proposed in this paper. This extraction is performed by parallel processes implementing an expert system represented by a grammar of frames.", | |
| "cite_spans": [ | |
| { | |
| "start": 31, | |
| "end": 41, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 42, | |
| "end": 63, | |
| "text": "AAA/ Conf., Aug. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 635, | |
| "end": 645, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 646, | |
| "end": 667, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1045, | |
| "end": 1055, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1056, | |
| "end": 1077, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Department of Computer Science Brown University", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Department of Computer Science The University of Rochester Rochester, New York 14627 Proc. 1982 AAAI Conf., Aug. 1982 The parsing of natural language is the product of dense interactions among various comprehension processes. We believe that traditional models have greatly underestimated the richness of these interactions. We propose a model for low-level parsing which is massively parallel, highly distributed, and highly connected. The model suggests a solution to the problem of word sense disambiguation which is psychologically plausible and computationally feasible.", | |
| "cite_spans": [ | |
| { | |
| "start": 85, | |
| "end": 95, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 96, | |
| "end": 117, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Steven Small, Gary Cottrell, and Lokendra Shastri", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The paper outlines the general connectionist paradigm followed by a brief description of a three-level network to do parsing. Finally we trace through an example to illustrate the functioning of the model. Proc. 1982 AAAI Conf., Aug. 1982 As children learn language they initially misunderstand reversible passive sentences as if they were active sentences. This error is an important clue to possible mechanisms by which children learn to understand passives in general. This paper reports on how the CHILD program learns to understand passive sentences, initially misunderstanding reversible passives as it does so. It presents an explanation of children's performance based on CHILD, and presents a number of predictions which follow from this explanation.", | |
| "cite_spans": [ | |
| { | |
| "start": 206, | |
| "end": 216, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 217, | |
| "end": 238, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Steven Small, Gary Cottrell, and Lokendra Shastri", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Champaign, Illinois 61820 Proc. 1982 AAAI Conf., Aug. 1982 This paper proposes that a comprehensive theory of stories must include components that deal with: (a) plan understanding, (b) narrative comprehension, and (c) the unique structural and affective aspects of the subclass of narratives that are stories. Proc. 1982 AAAI Conf., Aug. 1982 This paper presents a theory of AFFECT processing in the context of BORIS (Dyer, 1982) (Dyer, 1981a) , a computer program designed to read and answer questions about complex narratives.", | |
| "cite_spans": [ | |
| { | |
| "start": 26, | |
| "end": 36, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 37, | |
| "end": 58, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 311, | |
| "end": 321, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 322, | |
| "end": 343, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 418, | |
| "end": 430, | |
| "text": "(Dyer, 1982)", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 431, | |
| "end": 444, | |
| "text": "(Dyer, 1981a)", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "University of Illinois 603 East Daniel Street", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Here, \"complex\" entails the coordination, application, and search of many distinct sources of knowledge during both comprehension and question answering. This paper concentrates only on those structures and processes which interact with affect situations. The affect component in BORIS is not a separate module, but rather a series of structures and processes which arise as various lexical items are encountered during narrative comprehension and question answering (Dyer, 1981b ).", | |
| "cite_spans": [ | |
| { | |
| "start": 467, | |
| "end": 479, | |
| "text": "(Dyer, 1981b", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Affect Processing for", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Department of Computer Science Yale University P.O. Box 2158 (YS) New Haven, Connecticut 06520 Proc. 1982 AAA/ Conf., Aug. 1982 This paper discusses recognizing and producing contradictions. After illustrating the phenomena of contradiction, the paper presents conceptual classes of contradiction and gives an overview of how they can be recognized. The next part discusses the construction of contradictions, in particular with respect to contradicting historic events. The object of this paper is to examine the computational logic of contradictions, using contradictions as an example of how reasoning processes can and must exploit semantic knowledge and episodic memory, and to illustrate the kind of metaknowledge needed to use certain reasoning devices correctly and effectively.", | |
| "cite_spans": [ | |
| { | |
| "start": 95, | |
| "end": 105, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 106, | |
| "end": 127, | |
| "text": "AAA/ Conf., Aug. 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Margot Flowers", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Bill Swartout use/Information Sciences Institute 4676 Admiralty Way Marina del Rey, California 90291 Proc. 1982 AAAI Conf., Aug. 1982 This paper describes a prototype English generator which can produce English descriptions of program specifications written in Gist, a program specification language being developed at ISI. Such a facility is required because although Gist is a high level specification language, specifications written in it, like those in all other formal specification languages, are unreadable. There are several reasons for this unreadability: strange syntax; redundancy elimination; lack of thematic structure; implicit remote interactions; no representation of the motivation or rationale behind the specification; and a strict reliance on textual presentation. The current generator deals with the first two problems and part of the third. Our plans for dealing with the rest are outlined after a description of the current generator.", | |
| "cite_spans": [ | |
| { | |
| "start": 101, | |
| "end": 111, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 112, | |
| "end": 133, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "GIST English Generator", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Coordinated Science Laboratory University of Illinois Urbana, Illinois 61820 Proc. 1982 AAAI Conf., Aug. 1982 This paper outlines an approach to schema acquisition. The approach, called explanatory schema acquisition is applicable in problem solving situations and is heavily knowledge-based. Basically, learning is viewed as a fundamental part of the understanding process. Understanding a situation for which there is no existing schema involves generalizing the new event into a nascent schema. The new schema is then available to aid in future processing. This approach to learning is unique in several respects: it is not inductive and so is capable of one trial learning, it does not depend on failures to drive the learning process, and it is incremental and learns comparatively slowly. The learning procedure is outlined briefly with an example, a taxonomy of situations involving explanatory schema acquisition is given, and there is a brief discussion on the scope of the learning mechanism. Philadelphia, Pennsylvania 19104 Proc. 1982 AAAI Conf., Aug. 1982 This paper discusses the application of a propositional temporal logic to determining the competence of a monitor offer as an extended response by a question-answering system.", | |
| "cite_spans": [ | |
| { | |
| "start": 77, | |
| "end": 87, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 88, | |
| "end": 109, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1003, | |
| "end": 1046, | |
| "text": "Philadelphia, Pennsylvania 19104 Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1047, | |
| "end": 1068, | |
| "text": "AAAI Conf., Aug. 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Gerald DeJong", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Determining monitor competence involves reasoning about the possibility of some future state given a description of the current state and possible transitions.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Eric Mays Department of Computer and Information Science Moore School of Electrical Engineering/D2 University of Pennsylvania", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 50-53. The knowledge needed to process natural language comes from many sources. While the knowledge itself may be broken up modularly, into knowledge of syntax, semantics, etc., the actual processing should be completely integrated. This form of processing is not easily amenable to the type of processing done by serial \"yon Neumann\" computers. This work in progress is an investigation of the use of a spreading activation and lateral inhibition network as a mechanism for integrated natural language processing.", | |
| "cite_spans": [ | |
| { | |
| "start": 15, | |
| "end": 54, | |
| "text": "Conf. Cog. Sci. Soc., Aug. 1982, 50-53.", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 62-66. Newspaper cartoons can graphically display the results of ambiguity in human speech. The result can be unexpected and funny. Captioned cartoons derive their humor from a sudden incongruity which'can be made to follow by a human being who can automatically use stored world knowledge to resolve the ambiguous situation. Likewise, computer analysis of natural language statements also needs to successfully resolve ambiguous situations.", | |
| "cite_spans": [ | |
| { | |
| "start": 15, | |
| "end": 54, | |
| "text": "Conf. Cog. Sci. Soc., Aug. 1982, 62-66.", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Computerized understanding of dialogue that takes place between humans must not only include syntactical and semantical analysis, but also pragmatical analysis. Pragmatics consists of an understanding of the speaker's intentions, the context of the utterance, and social implications of polite human communication.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Computer techniques have already been developed to use restricted world knowledge in resolving ambiguous language use. This paper illustrates how these techniques can be used in resolving ambiguous situations arising in cartoons. Identifying factors that influence pronoun reference assignment is a challenge to anyone attempting to characterize the process of language understanding. Because a pronoun itself carries only a small part of the meaning that the understander is expected to assign to it, he or she must use contextual information to assign the pronoun an unambiguous referent. Characterizing aspects of the context which are used for this purpose is an active area of psychological research.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Many recent studies have considered the role of syntactic context, that is, the effect of structural constraints on pronoun reference in a fragment of text, typically a sentence, without recourse to constraints which might be found in the meaning of the text. Investigators have also examined the role of semantic factors within sentences in directing the assignment of referents. The studies reported here focus on the use of pragmatic constraints in resolving anaphoric pronouns.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "We are concerned with characterizing two major sources of contextual information in paragraphlength texts, and evaluating their influence on pronominal reference assignment.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 90-93. Inference can be viewed as a search through a space of inference rules. Backward and forward inference differ in the direction of the search: backward inference searches from goals to ground assertions; forward inference searches from ground assertions to goals. This paper describes an inference procedure, called bi-directional inference, which limits the number of inference rules searched. Bi-directional inference resuits from the interaction between forward and backward inference and loosely corresponds to bidirectional search. We show through an example that, when used throughout a session of related tasks, bidirectional inference sets up a conversational context and prunes the search through the space of inference rules by ignoring rules which are not relevant to that context. Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 146-148. Step-by-step natural language specification provides powerful intuitions for novice programmers using a programming language.", | |
| "cite_spans": [ | |
| { | |
| "start": 15, | |
| "end": 54, | |
| "text": "Conf. Cog. Sci. Soc., Aug. 1982, 90-93.", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 857, | |
| "end": 903, | |
| "text": "Ann. Conf. Cog. Sci. Soc., Aug. 1982, 146-148.", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "We hypothesize that these intuitions take the form of frame-like plans ~ regular but flexible techniques for specifying how to accomplish a task. Programming knowledge also involves frame-like plans. While an individual programming language plan may have many lexical and syntactic similarities to a corresponding natural language plan, the two plans often have incompatible semantics and pragmatics. Many novice programmers' misconceptions derive directly from these incompatibilities. In this brief report we show an example of natural language and programming language plans. Using those plans we discuss transcripts of novice programmers using a natural language plan while attempting a programming language problem.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "We conclude with a brief discussion of the implications of this work.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Natural Language Processing Using Spreading Activation and Lateral Inhibition Jordan Pollack and David Waltz Coordinated Science Laboratory University of Illinois Urbana, Illinois 61801", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 167-168. I have developed a question-answering program that will answer questions about simple stories. In my program, question-answering is divided into two processes:", | |
| "cite_spans": [ | |
| { | |
| "start": 10, | |
| "end": 56, | |
| "text": "Ann. Conf. Cog. Sci. Soc., Aug. 1982, 167-168.", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Department of EECS University of California, Berkeley Berkeley, California 94720", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "(1) answer formation and (2) answer expression. The program first looks down a causal chain which is formed by the story-understanding program and figures out in what part of the chain the answer lies. The answer can also be a subset of the chain, sometimes a quite long one. The second part of the program takes this long chain and decides what things are important to express to the questioner. This answer expresser uses general rules of expression to figtire out what it needs to include to make the answer understandable, informative and interesting.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Department of EECS University of California, Berkeley Berkeley, California 94720", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "This solution is different from other questionanswering algorithms which view question answering as one process. These programs gather possible answers, and then choose the 'best' answer from among them. My program works in conjunction with PAME-LA, a story-understanding program that specializes in goal-based stories. Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 169-173. In this short paper, we suggest a framework for the study of schematic aspects of natural language comprehension. The approach draws from previous work in schematic representation and reasoning, spreading activation, parsing, speech recognition, psycholinguistics, and computer vision. By decomposing schematic knowledge into diffuse units and by studying the way these facets of knowledge are connected inferentially, we expect to show important results in several areas. We use a particular spreading activation or active semantic network scheme, called connectionism, which consists of a massive number of appropriately connected computing units that communicate through weighted levels of excitation and inhibition. We show how a number of classical problems in the theory of schemata might be approached in a new way. Three principal issues are discussed: (1) comprehension takes place on a number of interacting levels of processing; (2) multiple hypotheses are simultaneously maintained at a number of diffuse processing loci; and 3 Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 174-176. This paper describes the language understanding component of the Unix Consultant (UC) system being developed at the Berkeley Artificial Intelligence Research project. The purpose of UC is to hold a conversation with a naive user of the Unix operating system while he or she is working on the computer, answering questions and solving problems for the user. The system has several other components, including the common sense planner PANDORA and the plan understander PAMELA. Our natural language understanding system contains as a subpart the PHRAN phrasal analysis program. The current system attempts to deal with the fact that PHRAN by itself is unable to deal with reference, and cannot disambiguate unless the linguistic patterns used require a particular semantic interpretation of the words. In addition, we wish to account for the fact that the same utterance may be interpreted differently in different contexts. noun phrases. In connection with ellipsis generation, anticipation of the way in which the user would be likely to reconstruct a given utterance can help to ensure that the system's utterances are not so brief as to be ambiguous or misleading. When generating noun phrases to characterize specific objects with which the user is not familiar, the system may take into account the existential assumptions, domain-related desires, and reference beliefs ascribed to the partner. These applications of user modelling are illustrated as realized in the dialogue system HAM-ANS, and some possible generalizations and extensions of the strategies described are discussed. Proc. 1982 European AI Conf., July 1982 This paper outlines a program that translates in the restricted context of cooking recipes from French into Arabic. It consists of two completely independent phases: the comprehension phase receives the French recipe text and outputs a sequence of elementary actions; the second phase generates from that sequence several texts all expressing, in Arabic, the same original recipe. Thus the translation is actually done text for text. The two phases are not fully developed here; but we describe a general scheme of the program and point out the analogy between them.", | |
| "cite_spans": [ | |
| { | |
| "start": 330, | |
| "end": 376, | |
| "text": "Ann. Conf. Cog. Sci. Soc., Aug. 1982, 169-173.", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1427, | |
| "end": 1473, | |
| "text": "Ann. Conf. Cog. Sci. Soc., Aug. 1982, 174-176.", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 3061, | |
| "end": 3071, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 3072, | |
| "end": 3100, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Department of EECS University of California, Berkeley Berkeley, California 94720", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Computer Laboratory University of Cambridge Corn Exchange Street CambridgeCB23QG, ENGLAND Proc. 1982 European AI Conf., July 1982 The aim of the work reported here is to maximize the use of general semantic information in an AI task processor, specifically in a system front end for converting natural language questions into formal database queries. The paper describes the translation component of such a front end, which is designed to work from the question meaning representation produced by a language analyser exploiting only general semantics and syntax, to a formal query relying on databasespecific semantics and syntax. Translation is effe-cted in three steps, and the paper suggests that the rich and explicit meaning representations using semantic primitives produced for input sentences by the analyser constitute a natural and effective base for further processing.", | |
| "cite_spans": [ | |
| { | |
| "start": 90, | |
| "end": 100, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 101, | |
| "end": 129, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Steps Towards Natural Language to Data Language Translation Using General Semantic Information B.K. Boguraev and K. Sparck Jones", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Computer Science Department University of New Hampshire Durham, New Hampshire 03824 Proc. 1982 European AI Conf., July 1982 This paper presents an approach to a computer analysis of simple metaphorical statements which provides a framework in which factors salient to the metaphor as well as conceptual information essential to the interpretation can be consistently incorporated. This approach considers both \"what is happening,\" i.e. predicative or explicit information, and the impression which the speaker/writer wishes to create, as represented by modifying or implicit information.", | |
| "cite_spans": [ | |
| { | |
| "start": 84, | |
| "end": 94, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 95, | |
| "end": 123, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Formalizing Factors in Metaphorical Extension Sylvia Weber Russell", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The predicative aspect provided the basis of a previously implemented system, which was based on abstracted attributes and relations, but also allowed for evaluative and magnitudinal factors; this system paraphrased statements including metaphorically used verbs. The proposed expanded system allows for the incorporation of effects or impressions on the participants of the metaphorically expressed conceptualization, on \"outside observers\" and, indirectly, on the hearer/reader. The concept of force, which is seen as frequently, structuring motivations for and conditions on changes of state, constitutes one of the factors shown to be useful in the analysis of such effects.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Formalizing Factors in Metaphorical Extension Sylvia Weber Russell", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Department of Applied Mathematics Faculty of Mathematics and Physics Charles University Malostranske n. 25 118 00 Prague 1, CZECHOSLOVAKIA Proc. 1982 European AI Conf., July 1982 Information systems with natural language access to stored data range from rather simple systems for specific purposes up to experimental projects of natural language understanding systems compiling also the data in an automatic way. Many of these projects use a trial-and-error approach; to make such a system more general, it seems advisable to include a relatively complete analysis of language structure; this ensures a general base, which can be simplified with respect to particular applications.", | |
| "cite_spans": [ | |
| { | |
| "start": 139, | |
| "end": 149, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 150, | |
| "end": 178, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Automatic Understanding with a Linguistically Based Knowledge Representation Petr Sgall", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Using operational tests for distinguishing between meaning and factual knowledge, ambiguity and indistinctness, etc., the Prague", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Automatic Understanding with a Linguistically Based Knowledge Representation Petr Sgall", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "American Journal of Computational Linguistics, Volume 8, Number 3-4, July-December 1982", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| } | |
| ], | |
| "back_matter": [ | |
| { | |
| "text": "The FINITE STRING Newsletter The FINITE STRING Newsletter", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "annex", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "The system we are currently constructing has a mechanism, called the context model which contains a record of knowledge relevant to the interpretation of the discourse, with associated levels of activation. There are rules governing how elements introduced into the context model are to influence it and the system's behavior. Proc. 4th Ann. Conf. Cog. Sci. Soc., Aug. 1982, 177-180. In the course of understanding a text, a succession of decision points arise at which readers are faced with the task of choosing among alternative possible interpretations of what they're reading. Careful analysis of a wide range of sample texts reveals that such decisions are often based on complex evaluations of the interpretation being constructed, and sometimes cause the reader to construct and discard a number of intermediate inferences before settling on a final interpretation for a text. This paper describes Judgmental Inference theory as a proposed scheme of evaluation metrics and mechanisms, derived from examination of inference decisions arising during text understanding. A series of programs, ARTHUR, MACARTHUR, and JUDGE are briefly described, which incorporate some of the metrics and mechanisms of Judgmental Inference, enabling them to understand texts more complex than those that can be handled by other understanding systems. Proc. 1982 European A/ Conf., July 1982 The process of discourse comprehension presupposes close co-operation of two subsystems of an understanding system --its knowledge (memory) subsystem and reasoning subsystem. The authors present a concrete mechanism called SCENE SURVEYOR for guiding reasoning in a text understanding system TAR-LUS. The SCENE SURVEYOR is an active structure which makes use of the knowledge of the world and the knowledge of the regularities of text construction.", | |
| "cite_spans": [ | |
| { | |
| "start": 337, | |
| "end": 383, | |
| "text": "Ann. Conf. Cog. Sci. Soc., Aug. 1982, 177-180.", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1338, | |
| "end": 1348, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1349, | |
| "end": 1377, | |
| "text": "European A/ Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Abstracts of Current Literature", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Its active character is manifested in posing questions about incoming data and looking for answers to these questions in order to recognize higher level situations (events not mentioned in the text explicitly), and secondly, in keeping track of scenes and their changes in the text. Proc. 1982 European AI Conf., July 1982 Efficient syllabic hypothesization in continuous speech has been so far an unsolved problem. A novel solution based on the extraction of acoustic cues is proposed in this paper. This extraction is performed by parallel processes implementing an expert system represented by a grammar of frames which is a generalization of an attributed grammar. Proc, 1982 European AI Conf., July 1982 An approach to parallel processing of natural language is described.It exploits the phenomenon of locality of text elements and is based on a multiprocessor system architecture which is aided by a special type of table grammars. The operation of the system is illustrated by an example of syntactic parsing in a subset of English.We describe the Friendly-Neighbours and Pyramid algorithms for parallel parsing.The FINITE STRING Newsletter Abstracts of Current Literature research group has formulated the level of meaning as a set of underlying representations consisting in dependency trees with nodes labelled by complex symbols, every lexical unit having its valency (case) frame. The relationships between this level of meaning and the domain of cognitive structures is handled by means of inference rules. The system described is being prepared for experiments in question-answering, based only on input texts and inference rules.", | |
| "cite_spans": [ | |
| { | |
| "start": 283, | |
| "end": 293, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 294, | |
| "end": 322, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 669, | |
| "end": 679, | |
| "text": "Proc, 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 680, | |
| "end": 708, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "On the Interaction of Knowledge", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Proc. 1982 European A/ Conf., Ju/y 1982, 244-249. Although a great deal of research effort has been expended in support of natural language (NL) database querying, little effort has gone into NL database update. One reason for this is that in NL querying one can tie nouns and stative verbs in the query to database objects (relation names, attributes and domain values).In many cases this correspondence seems sufficient to interpret NL queries. NL update seems to require database counterparts for active verbs, such as hire, schedule and enroll. We suggest a database counterpart for active verbs, which we call verbgraphs, that may be used to support NL update. A verbgraph is a structure for representing the various database changes that a given verb might describe. Oth.er possible uses of verbgraphs include specification of defaults, prompting of the user to guide user interaction, and enforcing database integrity constraints. Proc. 1982 European AI Conf., July 1982 Within a unique formalism, logic is a powerful theoretical basis allowing one to represent a database and to express the different types of processing --syntactic, semantic and deductive --involved when this database is consulted in natural language.In this framework we describe a complete system using a three-truth valued logic, rigorously defined. This logic allows a very fine representation of question semantics and lays the theoretical basis for the creation of an informative system consulted by casual or non-expert users. We compare the performance of our system with that of related ones, and outline possible extensions. Proc. 1982 European AI Conf., July 1982 This note is intended as a status report on our ongoing work in the area of natural language interaction with dynamic knowledge bases. Specifically, we discuss that portion of our research involving the offer by the question-answering system to monitor for future changes in the knowledge base.", | |
| "cite_spans": [ | |
| { | |
| "start": 30, | |
| "end": 49, | |
| "text": "Ju/y 1982, 244-249.", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 938, | |
| "end": 948, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 949, | |
| "end": 977, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1612, | |
| "end": 1622, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1623, | |
| "end": 1651, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "David Maier and Sharon C. Salveter Department of Computer Science State University of New York at Stony Brook Stony Brook, New York 11794", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Guenther Goerz ERZE Univ. Erlangen-Nuernberg Erlangen, W. GERMANY and Neuropsychiatric Institute University of California, Los Angeles Los Angeles, California Proc. 1982 European AI Conf., July 1982 We describe GLP, a chart parser that will be used as a SYNTAX module of the Erlangen Speech Understanding System. After a brief outline of the speech system's architecture we introduce the concept of chart parsing. The parser itself realizes a multiprocessing structure using an agenda, which easily allows the application of various parsing strategies in a transparent way. Finally we discuss which features have to be incorporated into the parser in order to process speech data: direction independent island parsing, handling of scores, and the interaction with higher level components like SEMANTICS. . 1982 European AI Conf., July 1982 A scenario for story writing is presented which attempts to improve that of TALE-SPIN. The overall model presented attempts to simulate Goal Directed Behavior of characters as well as intentions of the story written. The scenario of a computer program ROALD based on a sub-set of the model is presented.", | |
| "cite_spans": [ | |
| { | |
| "start": 159, | |
| "end": 169, | |
| "text": "Proc. 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 170, | |
| "end": 198, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 804, | |
| "end": 810, | |
| "text": ". 1982", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 811, | |
| "end": 839, | |
| "text": "European AI Conf., July 1982", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Applying a Chart Parser to Speech Understanding", | |
| "sec_num": null | |
| } | |
| ], | |
| "bib_entries": {}, | |
| "ref_entries": { | |
| "FIGREF1": { | |
| "text": "Abbott and John B. Black Cognitive Science Program Yale University New Haven, Connecticut 06520Proc. 4th Ann. Conf. Cog. Sci. Soc.,Aug. 1982, 70-72.", | |
| "uris": null, | |
| "num": null, | |
| "type_str": "figure" | |
| }, | |
| "FIGREF3": { | |
| "text": "context affects processing in both top-down and bottom-up directions.", | |
| "uris": null, | |
| "num": null, | |
| "type_str": "figure" | |
| }, | |
| "TABREF0": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table/>", | |
| "html": null | |
| }, | |
| "TABREF1": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table/>", | |
| "html": null | |
| }, | |
| "TABREF2": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table><tr><td>Department of EECS</td></tr><tr><td>University fo California, Berkeley</td></tr><tr><td>Berkeley, California 94720</td></tr></table>", | |
| "html": null | |
| }, | |
| "TABREF3": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table/>", | |
| "html": null | |
| }, | |
| "TABREF4": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table/>", | |
| "html": null | |
| }, | |
| "TABREF5": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table><tr><td>P.O. Box 2158</td></tr><tr><td>New Haven, Connecticut 06520</td></tr></table>", | |
| "html": null | |
| }, | |
| "TABREF6": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table/>", | |
| "html": null | |
| }, | |
| "TABREF7": { | |
| "text": "", | |
| "type_str": "table", | |
| "num": null, | |
| "content": "<table><tr><td>L.R.I. Bat 490</td></tr><tr><td>University of Paris South</td></tr><tr><td>91405 Orsay, FRANCE</td></tr></table>", | |
| "html": null | |
| } | |
| } | |
| } | |
| } |