Structural Non-Correspondence in Translation
What it is about
The paper examines the co-description approach of Kaplan et al. (1989) and shows that its most natural analysis of one translation phenomenon cannot cover the data. The phenomenon is the one where a source structure must be mapped onto a differently shaped target structure. The authors then go through alternatives: changing the correspondence notation, extending the formalism, or moving the exception to a different level. Their conclusion is that the central problem is to state the mapping in simple general terms while keeping the exceptional cases local, so that their effect does not spread through the grammar. They raise doubts about the feasibility of a large system on this basis. The paper has an implementation, but reports no figures.
Ideas
- I1. Measure locality of the deltas: when a delta is applied, count the tokens outside the concept it names that change. The target is zero. A delta that spreads is the paper's failure in measurable form.
- I2. Check functionality of the mapping: in the learned lexicon, count the many-to-one and one-to-many cases per concept. The paper's problem with a mapping that is not a function appears there as a count, and a gate can reject merges that break it.
- I3. Direction test: a correspondence that works one way (the fusing direction in the paper) may fail the other. Learn each Bible delta in both directions and report the residual of each, to see which exceptions are one-way.
For Mova
The code is tools/translate/src/meta.rs and measure.rs, where the delta and its residual are computed. The paper gives no method to copy; its useful content is the test of locality and functionality, which can be written as counts over the data.
Cite
@inproceedings{sadler1991structural,
title = {{Structural Non-Correspondence in Translation}},
author = {Sadler, Louisa and Thompson, Henry S.},
booktitle = {Fifth Conference of the European Chapter of the Association for Computational Linguistics},
year = {1991},
url = {https://aclanthology.org/E91-1051/}
}References
- I-lalvomen, P-K and R.M. Kaplan (1988) "Projections and Semantic Description in Lexical-Functional Grammar" presented at the International Conference on Fifth Generation Computer Systems, Tokyo, Japan. find
- Kaplan, R. (1987) "Three seductions of computational psycholingulstics", in P.Whitelock, M.M.Wood, H.L.Somers, RJolmson and P.Benuett (eds) Linguistic : Theory and Computer Applications, : Academic Press, London, pp 149-88. ~Kaplan, R., K. Netter, J. Wedekind and A. Zaenan (1989) "Translation by Structural Correspondences", Proceedings of 4th EACL, UMIST, Manchester, pp 272-81. find our card
- Sadler, L., I. Crookston and A. Way (1989) "Co-description, projection and 'difficult' translation", Working Papers in Language Processing No. 8, Dept. of Language and Linguistics, University of Essex. find
- Sadler, L., I. Crookston, D. Arnold and A. Way (1990) "LFG and Translation", in Third International Conference on Theoretical and Methodological Issues in Machine Translation, Linguistics Research Center, Austin, Texas, (Pages not numbered). van Noord, G., J.Dorrepaal, P. van der Eijk, M. Fiorenza, and L. des Tombe (1990) "The MiMo2 Research System" in Third International Conference on Theoretical and Methodological Issues tn Machine Translation, Linguistics Research Center, Austin, Texas, (Pages not numbered). find
- Wedeldnd, J. (1988) "Generation as structure driven derivation" Proceedings of COLING-88, vol 2, pp. 732-737, Budapest. find
- Zajac, R. (1990) "A Relational Approach to Translation" in Third International Conference on Theoretical and Methodological Issues in Machine Translation, Linguistics Research center, Texas, (Pages not numbered). Austin, - 29,0,. find
This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/sadler-1991-structural-non-correspondence-translation/