Reversible Unification Based Machine Translation

What it is about

The paper shows how bidirectional unification grammars can make a whole MT system reversible. Each language has one grammar that maps strings to language-dependent meaning representations and back, so the same grammar both parses and generates. Translation then runs through a chain of such grammars that meet at meaning representations. The author separates the reversible core, which produces the set of possible translations from linguistic knowledge alone, from the non-reversible components that choose the best translation from that set, which may use interaction or other knowledge. He notes that reversible systems were argued to give a better understanding of the translation relation and possibly better performance. The paper has no quantitative evaluation.

Ideas

I1. Split the translator into the core and the preference step, as the paper does. The core is the meta-world with both directions; the preference step is the author delta and the encoder's counts. Report the size of the candidate set and the rank of the reference inside it, separately from the exact-match score.
I2. Use one table for reading and writing where the data allows it: the chunk table is counted once and used both ways. Check with a roundtrip count that the encoder's table reproduces the reader's chunks for each concept.
I3. Enumerate the legal candidates for a verse, a small set, and count how many of them are legal translations of the same meta-world. The paper's split says this set is the thing to measure before any ranking is tuned.

For Mova

The code is tools/roundtrip (the tree and frame steps in both directions) and tools/translate (the meta-world as the shared core, profile.rs for the choice). The reversible-core idea gives Mova a clean measure of what the reader loses before any preference is applied.

Topics: Machine translation

Cite

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

@inproceedings{gertjan1990reversible,
  title = {{Reversible Unification Based Machine Translation}},
  author = {Gertjan, van Noord,},
  booktitle = {COLING 1990 Volume 2: Papers presented to the 13th International Conference on Computational Linguistics},
  year = {1990},
  url = {https://aclanthology.org/C90-2052/}
}

References

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

Douglas E. Appelt. Bidirectional gr~nmars and the design of natm'al language generation systems. In Theoretical Issues in Natural Language Processing 3, 1987. find
Jonathaal Calder, Mike Rcape, and tlenk Zeevat. An algorittma for generation in unification catcgorial grammar. In Fourth Con.ference of the European Chapter of the Association for Computational Linguistics, 1989. find
Marc Dymetman ~ld PielTe Isabellc. Reversible logic grannnars for machine translation. In Proceedings of the Second International Conference on Theoretical 3Although parsing of LFG's is decidable no such result is available for generation; note furthermore that according to [9] extension is Mlowed dm'ing generation. and Methodological issues in Machine Translation of Natural Languages, 1988. find
Barbara Qrosz, Karen Sparek Jones, and Bonny Lynn Webber, editors. Readings in Natural Language Processing. Morgan Kaufinann, 1986. find
Andrew Haas. A generalization of the offline parsable grammars. In '27th Annual Meeting oJ the Association for Computational Linguislies, 1989. find
Pien'e Isabelle. Towards reversible M T systems. In M T Summit ]I, 1989. find
Pien'e Isabelle, Marc Dymetman, and Elliott Macldo-vitch. CRITTER: a translation system for a~'ieulturM market reports. In Proceedings o/ the 12th International Conference on Computational Linguistics, Budapest, 1988. find
Paul S. Jacobs. Achieving bidirection'Mity. In Proceedings o] the 12th International Conference on Computational Linguistics, 1988. find
Ronald Kaplan, Klaus Netter, Jiirgen Wedekind, and Annie Zaenen. Translation by structural correspondences. In Fourth Conference o] the European Chapter of the Association for Computational Linguistics, 1989. find our card
Margaret King, editor. Machine Translation, the State o] the .4 ft. Edinburgh University Press, 1987. find
J a n Landsbergen. Isomorphic grammars and their use in the Rosetta translation system, 1984. paper presented at the tutorial on Machine Translation, Lugano t984, Also appears in [10]. find
Femando C.N, Pereira and David WaITen. Definite clause grammars for language analysis - a survey of the formMism and a comparison with augmented transition networks. Artificial Intelligence, 13, 1980. reprinted in 114]. find
Femando C.N. Pereira and David Wan'ca. Parsing as deduction. In 21st Annual Meeting of the Association /or Computational Linguistics, 1983. find
Carl Pollard and Ivan Sag. In.formation Based Syntax and Semantics. Center for the Study of Langalage and Information Stanford, 1987. find
Herbert Ruessink attd Gertjan van Noord, Remarks on the bottom-up generation Mgorithm. Technical report, Department of Linguistics, OTS RUU Utrecht, 1989. find
Stuart M. Stfieber. A mfiform architecture for parsing and generation. In Proceedings o/ the 12th International Conference on Computational Linguistics, 1988. find
Stuart M. SIfieber, Hans Uszkoreit, Fernando C.N. Pereira, J. Robinson, and M. Tyson. Tile formalism and implementation of PATR-II. In B. J. Grosz and M. E. Stickel, editors, Research on Interactive Acquisition and Use o] Knowledge. SRI report, 1983. find
Stuart M. Shieber, Gertjan van Noord, Robert C. Moore, and Fernando C.N. Pereira. A semantic-head-driven generation Mgorithm for unification based formMisms. In 27th Annual Meeting of the Association .for Computational Linguistics, 1989. find
Stuart M. Shieber, Gertjan van Noord, Robert C. Moore, and Fernando C.N. Pereira. Semantic-head-driven generation. Computational Linguistics, 1990. To appear. find
Tomek Strzalkowsld. Automated inversiot) of a unification parser into a unification generator. Technical report, Courant Institute of Mathematical Sciences, New York University, 1989. technical report 465. find
Gertjau van Noord. BUG: A directed lmttom-up generator for unification ba~sed fonnMisms. Working Pa-persia Natural Language Processing, Katholieke Universiteit Leuven, Stichting Taaltechnologie Utrecht, 4, 1989. find
Gertjan van Noord. An overview of head-driven bottom-up generation. In Robert Dale, Chris Mellish, and Michael Zock, editors, Current Research in Natural Language Generation. 1990. find
Jiirgen Wedekind. Generation as structure driven derivation. In Proceedings o/ the 12th International Conference on Computational Linguistics, 1988. find
Rdmi Zajac. A transfer model using a typed feature structure rewriting system with inheritcnce, In 27th Annual Meeting of the Association for Computational Linguistics, 1989. find
[leak Zeevat, Ewau Klein, and Jo Calder. Unification categorial grammar. In Nicholas tIaddock, Ewan Klein, and Glyn Morrill, editors, Categorial Grammar, Unification Grammar and Parsing. Centre for Cognitive Science, 1987. Volume 1 of Working Papel.'s in Cognitive Science. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/noord-1990-reversible-unification-based-machine-translation/