ABSent: Cross-Lingual Sentence Representation Mapping with Bidirectional GANs
Tuesday, March 03, 2020, 11:30am
Location : CBIM-22
Prof. Gerard de Melo (advisor), Prof. Karl Stratos, Prof. Yongfeng Zhang & Prof. Dong Deng (external committee member)
Event Type: Qualifying Exam
Abstract: A number of cross-lingual transfer learning approaches based on neural networks have been proposed for the case when large amounts of parallel text are at our disposal. However, in many real-world settings, the size of parallel annotated training data is restricted. Additionally, prior cross-lingual mapping research has mainly focused on the word level. This raises the question of whether such techniques can also be applied (weighted) averages of word vectors so as to effortlessly obtain cross-lingually aligned sentence representations. To this end, we propose an Adversarial Bi-directional Sentence Embedding Mapping framework, which learns mappings of cross-lingual sentence representations from limited quantities of parallel data. The experiments show that our method outperforms several technically more powerful approaches, especially under challenging low-resource circumstances.