Domain Mismatch Doesn’t Always Prevent Cross-lingual Transfer Learning
Noah A. Smith
Proceedings of the Thirteenth Language Resources and Evaluation Conference
Cross-lingual transfer learning without labeled target language data or parallel text has been surprisingly effective in zero-shot cross-lingual classification, question answering, unsupervised machine translation, etc. However, some recent publications have claimed that domain mismatch prevents cross-lingual transfer, and their results show that unsupervised bilingual lexicon induction (UBLI) and unsupervised neural machine translation (UNMT) do not work well when the underlying monolingual corpora come from different domains (e.g., French text from Wikipedia but English text from UN proceedings). In this work, we show how a simple initialization regimen can overcome much of the effect of domain mismatch in cross-lingual transfer. We pre-train word and contextual embeddings on the concatenated domain-mismatched corpora, and use these as initializations for three tasks: MUSE UBLI, UN Parallel UNMT, and the SemEval 2017 cross-lingual word similarity task. In all cases, our results challenge the conclusions of prior work by showing that proper initialization can recover a large portion of the losses incurred by domain mismatch.
Unsupervised Discovery of Firm-Level Variables in Earnings Call Transcript Embeddings
Proceedings of the Second Workshop on Financial Technology and Natural Language Processing
Compositional Morpheme Embeddings with Affixes as Functions and Stems as Arguments
Proceedings of the Workshop on the Relevance of Linguistic Structure in Neural Architectures for NLP
This work introduces a novel, linguistically motivated architecture for composing morphemes to derive word embeddings. The principal novelty in the work is to treat stems as vectors and affixes as functions over vectors. In this way, our model’s architecture more closely resembles the compositionality of morphemes in natural language. Such a model stands in opposition to models which treat morphemes uniformly, making no distinction between stem and affix. We run this new architecture on a dependency parsing task in Korean—a language rich in derivational morphology—and compare it against a lexical baseline,along with other sub-word architectures. StAffNet, the name of our architecture, shows competitive performance with the state-of-the-art on this task.