Shake-and-Bake Translation

What it is about

The paper proposes Shake-and-Bake translation, in which lexical items are multi-dimensional signs and translation equivalence is defined between bags (multisets) of lexical items. Source analysis gives the bag on one side, and generation in the target language combines the corresponding target signs freely according to the target grammar. Shared variables in the logical forms enforce the semantic constraints, so that very different syntactic and logical structures can be related. The author reviews the structure-based answers (transfer and interlingua) and argues that the bag approach keeps the grammars modular. The paper states that generation from a bag is NP-complete, so a tractable system needs more structure on the bag, and it suggests learning the bilingual correspondences from aligned corpora. An implemented trilingual English, French and Japanese system is mentioned, without figures.

Ideas

• I1. Treat the meta-world as a bag of concepts with relations, and let the encoder order it with the linearizer. Measure the search cost (candidate orders tried) on verses, and the exact-match rate, twice each.
• I2. Learn the bilingual correspondences from aligned verses, as the paper proposes. meta.rs already does this by counts; the test is the pair-trained word-by-word table against the meta-world on the same verses, which the translator's README reports as a tie on the distant pairs on test and a loss on dev.
• I3. Add an ordering prior: use the Markov classes in en/src/lin.rs to put the bag in a likely order before the search, and measure how many tries it takes to the first exact match. This is the paper's own fix for the NP-complete step, and it can be checked directly.

For Mova

The code is en/src/lin.rs (the ordering prior) and tools/translate (the meta-world as the bag of concepts). The bag view is a clean way to state the core of the meta-world; the search cost is what to measure before adding structure.

Topics: Machine translation

Cite

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

@inproceedings{whitelock1992shake,
  title = {{Shake-and-Bake Translation}},
  author = {Whitelock, P.},
  booktitle = {COLING 1992 Volume 2: The 14th International Conference on Computational Linguistics},
  year = {1992},
  url = {https://aclanthology.org/C92-2117/}
}

References

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

1. Allegranza, V. , P. Bennett, J. Durand, F. van Eynde, L. Humphries, P. Schmidt and E. Steiner (1991) "Linguistics for Machine Translation: The Eurotra Linguistic Specifications". Studies in MT and NLP, CEC. find
2. Alshawi, H. r D. M. Carter, B. Gambaeck and M. R a y n e r (1991) "Interactive Translation using Quasi-Logical Forms", Proceedings of the 29th ACrEs DE COL1NG-92. NANTES, 23-28 AOt~T 1992 ACL, Berkely. find
3. Baschung, K., G. Bes, A, Corluy and T. Guillotin (1987) "Auxiliaries and Clitics in French UCG Grammar", Proceedings of the Third European ACL, Copenhagen. find
4. Beaven, J.L. (1991) Lexicalist Unification-Based Machine Translation, PhD Thesis, submitted to Dept. of Artificial Edinburgh. Intelligence, Univ. of Brew, C.H. (1992) "Letting the cat out of the bag: Generation for Shake-and-Bake MT", this conference. find
5. Brown, P., J. Cocke, S. della Pietra, V.J. della Pietra, F. Jelinek, J.D.Lafferty, R.L. Mercer and P. S. Roossin (1990) "A Statistical Approach to Machine Translation", CL vol. 16, no 2. find
6. Calder, J. , M. Reape & H. Zeevat (1989l "An Algorithm for Generation in UCG", Proceedings of the Fourth European ACL, Manchester. find
7. Dowty~ D. (1989) "On the Semantic Content of the Notion 'Thematic-Role'", in G. Chierchia, B. Partee and R. Turner (eds.) Property Theory, Type Theory and Natural Language Semantics, Dordrecht, Reidel. find
8. Fenstad, J.E., P.K. Halvorsen, T. Langholm and J. van B e n t h a m (1985) Equations, Schemata and Situations: A Framework for Linguistic Semantics. CSLI-85-29, Stanford. find
9. Gunji, T. (1987) Japanese Phrase Structure Grammar, Reidel, Dordrecht. find
10. Hobbs, J. (1986) "Ontological Promiscuity", Proceedings of the 25th ACL. find
11. Kaplan, R. M. (1987) "Three Seductions in C o m p u t a t i o n a l P s y c h o l i n g u i s t i c s " , in P.J. Whitelock, M.M. Wood, H. Somers, P. Bennett, R. Johnson (eds.) Linguistic Theory and Computer Applications: Academic Press. find
12. Kaplan, R. M., K. Netter, J. Wedekind and A. Zaenen (1989) "Translation by Structural C o r r e s p o n d a n c e s " , Proceedings of the Fourth European ACL, Manchester. find
13. Karttunen, Lauri (1989) '~Radical Lexicalism", in M.R. Baltin and A.S. Kroch (eds.), Alternative Conceptions of Phrase Structure, Chicago. find
14. Landsbergen, J. (1987) "Montagne Grammar and Machine Translation", in P.J. Whitelock, M.M. Wood, H. Somers, P. Bem~ett, R. Johnson (eds.) Linguistic Theory and Computer Applications: PROC. OF COLING-92. NANTES, AUtL 23-28. 1992 Academic Press. find
15. Lytinen, S and R. Shank (1982) "Representation and Translation". TR 324, Dept. of Computer Science, Yale University. find
16. Moens, M., I. Calder, E. Klein, M. Reape and H. Zeevat (1989) "Expressing Generalisations in Unification-Based Grammar Formalisms", Proceedings of the Fourth European ACL, Manchester. find
17. Moortgat, M. (1988) Categorial Investigations: Logical and Linguistic Aspects of the Lambek Calculus. Foris, Dordrecht. find
18. Nagao, M. J. Tsujii and J. Nakamura (1985) 'q'he Japanese G o v e r n m e n t Project for Machine Translation", Computational Linguistics, vol. 11 #2-3 find
19. Odijk, J. (1989) '"I'he Organisation of the Rosetta Grammars", Proceedings of the 4th European ACL, Manchester. find
20. Parsons, T. (1980) "Modifiers and Quantifiers in N a t u r a l L a n g u a g e " , Canadian Journal of Philosophy, supp. vol. VI. find
21. Parsons, T. (1991) Events in the Semantics of English, MIT Press, Cambridge, Mass. find
22. Pereira, F. C.N. (1987) "Grammars and Logics of Partial Information", in Proceedings of the 4th International Conference on Logic Programming, Melbourne, Australia. find
23. Pollard, C and 1. Sag (1987)Information-based Syntax and Semantics: Vol 1: Fundamentals, CSLI Lecture Notes 13. find
25. Ramsay, A (1991) "A common framework for analysis and generation", in Proceedings of the 5th European ACL, Berlin. find
26. R~sner, D. (1986) "When Mariko talks to Siegfried: Experiences from a Japanese/German MT project". Proceedings of the 11th International Conference in Computational Linguistics (COLING), Bonn. find
27. Sadler, L., I. Crookston and A. Way (1989) "Co-description, projection and 'difficult' translation", Working Papers in Language Processing #8, Dept. of Language and Linguistics, University of Essex. find
28. Sadler, L., I. Crookston, D. Arnold and A. Way (1990) "LFG and Translation", in Third Ac'u~s DE COLING-92. NANT1~S. 23-28 hotrr 1992 International Conference on Theoretical and Methodological Issues in MT, Linguistics Research Centre, Austin, Texas. find
29. Sadler, L and H. S. Thompson (1991) "Structural N o n - C o r r e s p o n d a n c e in Translation", in Proceedings of the 5th European ACL, Berlin. find
30. S.M. (1986) An Introduction to Unification -Based Approaches to Grammar, Sheiber, University of Chicago Press. find
31. Shieber, S. M. (1988) A Uniform Architecture for Parsing and Generation", Proceedings of the 12th International Conference in Computational Linguistics (COLING), Budapest. find
32. Tsujti, J. (1986) "Future Directions of Machine Translation", Proceedings of the 11th International Conference in Linguistics (COLING), Bonn. find
33. Computational Tsujii, J and K. Fujita (1991) "Lexieal Transfer based on Bilingual Signs", Proceedings of the 5th European ACL, Berlin, April 1991. find
34. Ushida, H and K. Sugiyama (1980) "A Machine Translation system from Japanese into English based on Conceptual Structure". Proceedings of the 8th International Conference in Computational Linguistics (COLING), Tokyo. find
35. Uszkoreit, Hans (1986) "Categorial Unification Grammars", Proceedings of the 11th International Conference in Computational Linguistics (COLING), Bonn. find
36. van de Veen, E. (1990) Parsing Free Word Order Languages, MSc Thesis, Dept. of Artificial Intelligence, University of Edinburgh. find
37. Vauquois, B. and Ch. Boitet (1985) "Automated Translation at Grenoble University", Computational Linguistics, vol. 11 #1. find
38. W h i t e l o c k , P. (1991a) "Shake-and-Bake Translation". in Proceedings of the Workshop on Constraint Propagation and Linguistic Description. ed. C.J. Rupp, M. Rosner and R. Johnson, IDSIA, Lugano. find our card
39. Whitelock, P. (1991b)A Lexicalist Unification Grammar of Japanese. PhD Thesis, submitted to Dept. of Language and Linguistics, UMIST. find
40. Zeevat, H., E. Klein and J. Calder (1987) "An Introduction to Unification Calegorial Grammar", in N.J. Haddock, E. Klein and G. Morrill (eds.) Edinburgh Working Papers in Cognitive Science, vol. 1: Categorial Grammar, Unification Grammar and Parsing. Centre for Cog. Science, Edinburgh. find
41. PROC. or: COLING-92. NANTES. AUG. 23-28. 1992 SHAKE-AND-BAKE TRANSLATION find our card

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/whitelock-1992-shake-bake-translation/