Elamur Research⌕
☰
home / papers / 1996 / Anaphor Resolution and the Scope of Syntactic Constraints

Anaphor Resolution and the Scope of Syntactic Constraints

Summary

The paper asks how Chomsky's Binding Theory can be turned into a practical component of an anaphor resolver, and where purely syntactic constraints stop being enough. Stuckardt describes the resolver built into the KONTEXT text analysis system for German. It handles reflexives, nonreflexive pronouns and definite common nouns. The system first filters candidates using morphosyntactic agreement, binding principles A/B/C, and a definiteness condition on cataphora. It then ranks the remaining candidates with preferences such as recency, subject preference and case role inertia.

Resolving each anaphor independently can produce index assignments that jointly violate binding principles. A literal generate-and-test implementation avoids this but costs exponential time. The proposed algorithm instead orders decisions by plausibility and re-checks binding constraints dynamically against earlier choices. This catches such conflicts at modest cost, although it does not guarantee a globally optimal solution.

The author also shows that deverbal NPs can set up local binding domains whose implicit subject only pragmatic inference can identify. Syntax alone therefore cannot always decide binding.

Key points

  • Earlier work motivates the syntax-centred approach. Hobbs (1978) resolved over 80% of pronouns with a "naive" syntactic algorithm, and selectional constraints added about 3.5%. In Lappin and Leass (1994), lexical preference statistics added only about 3% and scored below 35% when used alone.
  • KONTEXT relies on a dependency grammar. Its dependency trees are converted into a simplified surface phrase structure, because binding-relevant details such as subject–object asymmetry and local domains are not visible in raw dependencies. The paper criticizes Strube and Hahn (1995) for checking binding directly on dependency structure.
  • The algorithm works in three steps:
    • It ranks candidates for each anaphor.
    • It sorts all anaphors globally by the plausibility of their best candidate.
    • It accepts a candidate only if agreement and binding still hold given the coindexings already chosen.
  • Contrasts like "accepts/revises the decision for him(self)" show that the implicit subject of a deverbal NP may require pragmatic inference. Approaches that consider only domains created by possessive determiners can therefore go wrong.
  • For future work, the author proposes integrating parsing decisions (attachment, syntactic function) with anaphor resolution rather than running them in sequence.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics: Coreference resolution

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{stuckardt1996anaphor,
  title = {{Anaphor Resolution and the Scope of Syntactic Constraints}},
  author = {Stuckardt, Roland},
  booktitle = {COLING 1996 Volume 2: The 16th International Conference on Computational Linguistics},
  year = {1996},
  doi = {10.3115/993268.993330},
  url = {https://doi.org/10.3115/993268.993330}
}

References

17 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Jaimc G. Carbonell, Ralf D. Brown. 1988. Anaphora Resolution: A Multi-Strategy Approach. In: Proc. I2th Int. Conf. on Computational Linguistics, COLING, Vol. I, 96-101, Budapest. find our card
  2. David Carter. 1987. Interpreting Anaphors in Natural Language Texts. Ellis Horwood Series in Artificial Intelligence, Chichester. find
  3. Noam Chomsky. 1981. Lectures on Government and Binding. Foris Publications, Dordrecht. find
  4. Noam Choinsky. 1986. Knowledge of Language. Praeger, New York. find
  5. Nelson Correa. 1988. A Binding Rule ]or Government-binding Parsing. In: Proc. 12th Int. Conf. on Computational Linguistics, COLING, Vol. I, 123-129, Budapest. find
  6. Barbara J. Grosz, Candace L. Sidner. 1986. Attention, intention and the Structure of Discourse, in: Computational Linguistics, Vol. 12, Number 3, 175-204. find
  7. Karin Haenelt. 1994. Das Textanalysesystem KONTEXT. Konzcption und AnwendungsmSg~ lichkciten. In: Sprache und Datcnvcrarbeitung, 18, 1, 17-31, 1994. find
  8. Jerry Hobbs. 1978. Resolving Pronoun Refer-enccs. In: Lingua, 44,311-338. find
  9. Susmnu Kuno. 1987. Anapho~t and Discourse Principles. In: Makoto Nagao (ed). find
  10. Language and Artificial Intelligence. Elsevier, 1987. find
  11. Vittorio M. Lampugnani (ed). 1983. Lexikon der Architektur des 20. ,Iahrhundcrts. Hatje, Stuttgart. find
  12. Shalom Lappin and Herbert J. Lea~s. 1994. An Algorithm for P~vnominal Anaphora Resolution. In: Computational Linguistics, 20 (4), 535-561. find
  13. Candace L. Sidner. 1983 I,bcusing in the Comprehension of Definite Anaphora. In: Michael Brady, Robert C. Berwick (eds). Computational Models of Discourse. M.I.T. Press, Cambridge, MA, 267-330. find
  14. Michael Strube, Udo Hahn. 1995. ParscTalk about Sentence- and Text-Level Anaphora. In: Proc. ~th Conf. of the European Chapter of the Association for Computational Linguistics, EACL, 237-244. Dublin, March 28-31, 1995. find
  15. Roland Stucl~trdt. 1996. An Interdependency-Sensitive Approach to Anaphor Resolution. To appear in: Proc. of DAARCg6 - Discourse Anaphora and Anaphor Resolu, tion Colloquium,. Lancaster University, 17-18th July, 1996. find
  16. Wolfgang Teubert. 1 9 7 9 . Valenz des Sub-stantivs. Attributive Ergiinzungen nnd Angaben. P~idagogischer Verlag Schwann. find
  17. Bonnie L. Webber. 1989. Deictic Reference and Discourse Structure. Department of Computer and Information Science, University of Pennsylvania, Philadelphia PA 19104-6389. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/stuckardt-1996-anaphor-resolution-scope-syntactic-constraints/