Discourse Analysis via Questions and Answers: Parsing Dependency Structures of Questions Under Discussion

被引:0
|
作者
Ko, Wei-Jen [1 ]
Wu, Yating [2 ]
Dalton, Cutter [4 ]
Srinivas, Dananjay [4 ]
Durrett, Greg [1 ]
Li, Junyi Jessy [3 ]
机构
[1] Univ Texas Austin, Comp Sci, Austin, TX 78712 USA
[2] Univ Texas Austin, Elect & Comp Engn, Austin, TX 78712 USA
[3] Univ Texas Austin, Linguist, Austin, TX 78712 USA
[4] Univ Colorado Boulder, Linguist, Boulder, CO USA
来源
FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2023) | 2023年
关键词
COHERENCE;
D O I
暂无
中图分类号
学科分类号
摘要
Automatic discourse processing is bottlenecked by data: current discourse formalisms pose highly demanding annotation tasks involving large taxonomies of discourse relations, making them inaccessible to lay annotators. This work instead adopts the linguistic framework of Questions Under Discussion (QUD) for discourse analysis and seeks to derive QUD structures automatically. QUD views each sentence as an answer to a question triggered in prior context; thus, we characterize relationships between sentences as free-form questions, in contrast to exhaustive fine-grained taxonomies. We develop the first-of-its-kind QUD parser that derives a dependency structure of questions over full documents, trained using a large, crowdsourced question-answering dataset DCQA (Ko et al., 2022). Human evaluation results show that QUD dependency parsing is possible for language models trained with this crowdsourced, generalizable annotation scheme. We illustrate how our QUD structure is distinct from RST trees, and demonstrate the utility of QUD analysis in the context of document simplification. Our findings show that QUD parsing is an appealing alternative for automatic discourse processing.
引用
收藏
页码:11181 / 11195
页数:15
相关论文
共 50 条
  • [41] Dentistry website analysis: An overview of the content of formulated questions and answers
    Bastani, Peivand
    Niknam, Fatemeh
    Rezazadeh, Mahboobeh
    Rossi-Fedele, Giampiero
    Edirippulige, Sisira
    Samadbeik, Mahnaz
    HELIYON, 2022, 8 (08)
  • [42] Psychiatric and injury prevention approaches to youth suicide: Questions, answers, and lethal means - Discussion
    Christoffel, KK
    JOURNAL OF THE AMERICAN ACADEMY OF CHILD AND ADOLESCENT PSYCHIATRY, 1998, 37 (05): : 481 - 484
  • [43] AN UNNOTICED READING FOR WH-QUESTIONS: ELIDED ANSWERS AND WEAK ISLANDS Squibs and Discussion
    Spector, Benjamin
    LINGUISTIC INQUIRY, 2008, 39 (04) : 677 - 686
  • [44] How Do We Answer Complex Questions: Discourse Structure of Long-form Answers
    Xu, Fangyuan
    Li, Junyi Jessy
    Choi, Eunsol
    PROCEEDINGS OF THE 60TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), VOL 1: (LONG PAPERS), 2022, : 3556 - 3572
  • [45] Court Standards for Terminating Employees Under the ADA: Some Questions and Answers for Managers
    Lawrence S. Kleiman
    David W. Denton
    Employee Responsibilities and Rights Journal, 2000, 12 (2) : 63 - 77
  • [46] Achieving Success Under Payment Reform-More Questions Than Answers
    Barnett, Michael L.
    Landon, Bruce E.
    JAMA NETWORK OPEN, 2019, 2 (07)
  • [47] Court standards for terminating employees under the ADA: Some questions and answers for managers
    Kleiman L.S.
    Denton D.W.
    Employee Responsibilities and Rights Journal, 2000, 12 (2) : 63 - 77
  • [48] MORE QUESTIONS AND ANSWERS ON COMPUTER MOLD FILLING ANALYSIS .2.
    NAITOVE, MH
    PLASTICS TECHNOLOGY, 1984, 30 (05) : 75 - 78
  • [49] Bridging the Gap between Synthetic and Natural Questions via Sentence Decomposition for Semantic Parsing
    Niu, Yilin
    Huang, Fei
    Liu, Wei
    Cui, Jianwei
    Wang, Bin
    Huang, Minlie
    TRANSACTIONS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, 2023, 11 : 367 - 383
  • [50] Place Questions and Human-Generated Answers: A Data Analysis Approach
    Hamzei, Ehsan
    Li, Haonan
    Vasardani, Maria
    Baldwin, Timothy
    Winter, Stephan
    Tomko, Martin
    GEOSPATIAL TECHNOLOGIES FOR LOCAL AND REGIONAL DEVELOPMENT, 2020, : 3 - 19