Machine Translation

Historical notes

1. What was the ALPAC report, and what was its effect?
   A critical 1966 US report (Automatic Language Processing Advisory Committee) that led to dramatic funding cuts for MT and a call for basic research in computational linguistics; it also included an influential early evaluation study.
   • Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 13, Historical notes, p. 298
2. What are the three classic MT architectures shown in the Vauquois triangle?
   Direct translation (word by word with a large bilingual dictionary), transfer (parse the source, transform the parse with rules, generate), and interlingua (analyze into an abstract meaning representation, generate from it). Deeper analysis needs less transfer knowledge.

   • Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 13, Historical notes, p. 299
3. How can the neural encoder-decoder be seen in terms of the Vauquois triangle?
   As an interlingual approach, with attention integrating direct and transfer approaches by letting the decoder access source words and their representations directly.
   • Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 13, Historical notes, p. 299
4. What drove statistical MT around 1990, and what were the IBM Models?
   Large bilingual corpora like the Canadian Hansards and then the web. The IBM group, drawing on the noisy channel model of speech recognition, proposed generative IBM Models 1–5 (the Candide system), published in full detail.
   • Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 13, Historical notes, pp. 299–300
5. What were phrase-based translation and MERT?
   Phrase-based translation: a generative statistical approach that induces translations of phrase pairs. MERT (Minimum Error Rate Training): directly optimizing a discriminative log-linear model for metrics like BLEU. Toolkits: GIZA, Moses.
   • Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 13, Historical notes, p. 300
6. What are transduction (synchronous) grammars?
   Grammars assigning parallel syntactic trees to a sentence pair, translating via reordering operations on the trees (e.g. inversion transduction grammar, synchronous CFGs).
   • Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 13, Historical notes, p. 300
7. Who pioneered neural encoder-decoder MT?
   Kalchbrenner & Blunsom (2013, CNN encoder + RNN decoder); first applied to MT by Bahdanau et al. (2015); the transformer encoder-decoder by Vaswani et al. (2017).
   • Jurafsky & Martin, Speech and Language Processing, 3rd ed. draft, ch. 13, Historical notes, p. 300