Anssi Yli-Jyrä

Anssi Yli-Jyrä, Josi Purhonen, Matti Liljeqvist et al.

Twenty-five years ago, morphologically aligned Hebrew-Finnish and Greek-Finnish bitexts (texts accompanied by a translation) were constructed manually in order to create an analytical concordance (Luoto et al., 1997) for a Finnish Bible translation. The creators of the bitexts recently secured the publisher's permission to release its fine-grained alignment, but the alignment was still dependent on proprietary, third-party resources such as a copyrighted text edition and proprietary morphological analyses of the source texts. In this paper, we describe a nontrivial editorial process starting from the creation of the original one-purpose database and ending with its reconstruction using only freely available text editions and annotations. This process produced an openly available dataset that contains (i) the source texts and their translations, (ii) the morphological analyses, (iii) the cross-lingual morpheme alignments.

Aarne Talman, Anssi Yli-Jyrä, Jörg Tiedemann

Sentence-level representations are necessary for various NLP tasks. Recurrent neural networks have proven to be very effective in learning distributed representations and can be trained efficiently on natural language inference tasks. We build on top of one such model and propose a hierarchy of BiLSTM and max pooling layers that implements an iterative refinement strategy and yields state of the art results on the SciTail dataset as well as strong results for SNLI and MultiNLI. We can show that the sentence embeddings learned in this way can be utilized in a wide variety of transfer learning tasks, outperforming InferSent on 7 out of 10 and SkipThought on 8 out of 9 SentEval sentence embedding evaluation tasks. Furthermore, our model beats the InferSent model in 8 out of 10 recently published SentEval probing tasks designed to evaluate sentence embeddings' ability to capture some of the important linguistic properties of sentences.

Anssi Yli-Jyrä

Sequential Constraint Grammar (SCG) (Karlsson, 1990) and its extensions have lacked clear connections to formal language theory. The purpose of this article is to lay a foundation for these connections by simplifying the definition of strings processed by the grammar and by showing that Nonmonotonic SCG is undecidable and that derivations similar to the Generative Phonology exist. The current investigations propose resource bounds that restrict the generative power of SCG to a subset of context sensitive languages and present a strong finite-state condition for grammars as wholes. We show that a grammar is equivalent to a finite-state transducer if it is implemented with a Turing machine that runs in o(n log n) time. This condition opens new finite-state hypotheses and avenues for deeper analysis of SCG instances in the way inspired by Finite-State Phonology.

Anssi Yli-Jyrä, Carlos Gómez-Rodríguez

We present a simple encoding for unlabeled noncrossing graphs and show how its latent counterpart helps us to represent several families of directed and undirected graphs used in syntactic and semantic parsing of natural language as context-free languages. The families are separated purely on the basis of forbidden patterns in latent encoding, eliminating the need to differentiate the families of non-crossing graphs in inference algorithms: one algorithm works for all when the search space can be controlled in parser input.