Syntactic and Semantic Transfer with F-Structures

What it is about

The paper gives two ways to transfer between languages from LFG f-structures, the attribute-value syntax of LFG. In the first, f-structures are translated into a term language and transferred there. In the second, f-structures are translated into Underspecified Discourse Representation Structures (UDRS), and transfer runs on those semantic structures inside the Verbmobil transfer module. The authors compare both with two older approaches, co-description and the restriction operator, and find the older ones hard to use for embedded head switching and for several adjuncts. Transfer on the semantic layer avoids those problems, because head switching is handled when the semantic representation is built. The semantic layer also keeps scope ambiguity open, so ambiguity-preserving transfer is possible. The rules are bidirectional, and the grammar needs no extra annotation of the lexicons. The parts of the paper read contain examples but no counts.

Ideas

• I1. Keep the attachment of adverbials open in the meta-world. Where a Bible verse has an adverbial with two plausible heads, keep both on the meta token, and let the encoder choose the one its standard supports. Count the verses with such adverbials in two translations before any change.
• I2. Read the meta-world from frames rather than from the tree. The paper's argument is that a semantic layer avoids the head-switch mismatch of syntactic transfer. Mova's open problem 1 names the same step. Measure attachment noise on the same pairs before and after, twice.
• I3. Keep the standard bidirectional. The paper's rules run both ways, so one table serves the reader and the encoder. Check round-trip failures per class (meta to text to meta) to see whether the tables really are symmetric.
• I4. Ask where each piece of work belongs, as the paper does: in the grammar, in the transfer or in the semantics. For each class of the residual, record the layer that is responsible.

For Mova

The places are tools/translate/meta.rs (the reader), tools/roundtrip (the frames) and the encoder profile.rs. I1 and I2 are the attachment problem of open problem 1; I3 is the symmetry of the standard. The paper gives no counts in the parts read, so every measure must be taken on Mova's data.

Topics: Machine translation

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.3115/980845.980903

@inproceedings{dorna1998syntactic,
  title = {{Syntactic and Semantic Transfer with F-Structures}},
  author = {Dorna, Michael and Frank, Anette and Genabith, Josef van and Emele, Martin C.},
  booktitle = {36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics, Volume 1},
  year = {1998},
  doi = {10.3115/980845.980903},
  url = {https://doi.org/10.3115/980845.980903}
}

References

16 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. H. Alshawi and R. Crouch. 1992. Monotonic semantic interpretation. In Proceedings of A CL, pages 32-39, Newark, Delaware. J. Bos, B. Gamb~ick, C. Lieske, Y. Mori, M. Pinkal, and K. Worm. 1996. Compositional Semantics in Verbmobil. Coling'96, pages 131-136, Copenhagen, Denmark. find
2. R. Caspari and L. Schmid. 1994. Parsing und Generierung in TrUG. Verbmobil Report 40, Siemens AG, December. M: Dalrymple, J. Lamping, F.C.N Pereira, and V. Saraswat. 1996. A deductive account of quantification in lfg. In M. Kanazawa, C. Pinon, and H. de Swart, editors, Quantifiers, Deduction and Context, pages 33-57. CSLI Publications, No. 57. find
3. M. Dorna and M. C. Emele. 1996a. Efficient Implementation of a Semantic-based Transfer Approach. ECAI'96, Budapest, Hungary. find
4. M. Dorna and M. C. Emele. 1996b. Semantic-based Transfer. Coling'96, Copenhagen, Denmark. find
5. M. C. Emele and M. Dorna. 1998. Ambiguity Preserving Transfer Using Packed Representations. Coling'98, Montreal, Canada. find
6. J. van Genabith and R. Crouch. 1996. Direct and underspecified interpretations of lfg f-structures. In COLING 96, Copenhagen, Denmark, pages 262-267. find
7. J. van Genabith and R. Crouch. 1997. On interpreting f-structures as udrss. In ACL-EACL-97, Madrid, Spain, pages 402-409. find
8. J. van Genabith, A. Frank, and M. Dorna. 1998. Transfer Constructors. LFG Conference '98, Brisbane, Australia. find
9. M. Johnson. 1991. Features and Formulae. Computational Linguistics, 17(2):131-151. find
10. R. M. Kaplan and J. Wedekind. 1993. Restriction and Correspondance-based Translation. EACL'93, pages 193-202, Utrecht, The Netherlands. find
11. R. Kaplan, K. Netter, J. Wedekind, and A. Zaenen. 1989. Translation by Structural Correspondences. EACL'8g, pages 272-281, Manchester, UK. find our card
12. M. Kay, M. Gawron, and P. Norwig. 1994. Verbmobil: a Translation System for Face-to-Face Dialogs. Number 33 in CSLI Lecture Notes. University of Chicago Press. find
13. John T. Maxwell III and Ronald M. Kaplan. 1993. The interface between phrasal and functional constraints. Computational Linguistics, 19(4):571-590. find
14. T. Parsons. 1991. Events in the Semantics of English. MIT Press, Cambridge, Mass. find
15. U. Reyle. 1993. Dealing with Ambiguities by Underspecification: Construction, Representation and Deduction. Jounal of Semantics, 10(2):123-179. find
16. L. Sadler and H. S. Thompson. 1991. Structural Non-correspondence in Translation. EACL'91, pages 293-298, Berlin, Germany. find our card

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/dorna-1998-syntactic-semantic-transfer-f-structures/