The large attention-based encoder-decoder network (Transformer) has become prevailing recently due to its effectiveness. But the high computation complexity of its decoder raises the inefficiency issue. By examining the mathematic formulation of the decoder, we show that under some mild conditions, the architecture could be simplified by compressing its sub-layers, the basic building block of Transformer, and achieves a higher parallelism. We thereby propose Compressed Attention Network, whose decoder layer consists of only one sub-layer instead of three. Extensive experiments on 14 WMT machine translation tasks show that our model is 1.42x faster with performance on par with a strong baseline. This strong baseline is already 2x faster than the widely used standard baseline without loss in performance.

Subword segmenters like BPE operate as a preprocessing step in neural machine translation and other (conditional) language models. They are applied to datasets before training, so translation or text generation quality relies on the quality of segmentations. We propose a departure from this paradigm, called subword segmental machine translation (SSMT). SSMT unifies subword segmentation and MT in a single trainable model. It learns to segment target sentence words while jointly learning to generate target sentences. To use SSMT during inference we propose dynamic decoding, a text generation algorithm that adapts segmentations as it generates translations. Experiments across 6 translation directions show that SSMT improves chrF scores for morphologically rich agglutinative languages. Gains are strongest in the very low-resource scenario. SSMT also learns subwords that are closer to morphemes compared to baselines and proves more robust on a test set constructed for evaluating morphological compositional generalisation.

Pretrained character-level language models were recently shown to be competitive with popular subword models across a range of NLP tasks. However, there has been little research on their effectiveness for neural machine translation (NMT). This work performs an extensive comparison across multiple languages and experimental conditions of state-of-the-art character- and subword-level pre-trained models (ByT5 and mT5, respectively) on NMT, showing the effectiveness of character-level modeling in translation, particularly in cases where training data is limited. In our analysis, we show how character models' performance gains are reflected in better translations of orthographically similar words and rare words. While evaluating the importance of source texts in driving model predictions, we highlight ByT5 word-level patterns suggesting an ability to modulate word and character-level information during the translation, providing insights into a potential weakness of character-level modeling. We conclude by assessing the efficiency tradeoff of character models, suggesting their usage in non-time-critical scenarios to boost translation quality.

Kurdish Sign Language (KuSL) is the natural language of the Kurdish Deaf people. We work on automatic translation between spoken Kurdish and KuSL. Sign languages evolve rapidly and follow grammatical rules that differ from spoken languages. Consequently,those differences should be considered during any translation. We proposed an avatar-based automatic translation of Kurdish texts in the Sorani (Central Kurdish) dialect into the Kurdish Sign language. We developed the first parallel corpora for that pair that we use to train a Statistical Machine Translation (SMT) engine. We tested the outcome understandability and evaluated it using the Bilingual Evaluation Understudy (BLEU). Results showed 53.8% accuracy. Compared to the previous experiments in the field, the result is considerably high. We suspect the reason to be the similarity between the structure of the two pairs. We plan to make the resources publicly available under CC BY-NC-SA 4.0 license on the Kurdish-BLARK (https://kurdishblark.github.io/).

We introduce a novel architecture, the Neuromodulation Gated Transformer (NGT), which implements neuromodulation in transformers via a multiplicative effect. We compare it to baselines and show that it results in the best average performance on the SuperGLUE benchmark validation sets. Cellular neuromodulation is a biological mechanism involving neurons, where their intrinsic properties are continuously modified in a context-dependent manner according to stimuli, i.e., biochemicals called neuromodulators (Bargmann & Marder, 2013; Marder et al., 2014; Shine et al., 2021; Vecoven et al., 2020); it allows for the regulation of a population of neurons (Katz & Edwards, 1999). It has achieved notable success in the continual learning domain (Beaulieu et al., 2020; Ellefsen et al., 2015; Velez & Clune, 2017). Transformers (Vaswani et al., 2017) are architectures that eliminate recurrence by relying entirely on attention.

The study of the attention mechanism has sparked interest in many fields, such as language modeling and machine translation. Although its patterns have been exploited to perform different tasks, from neural network understanding to textual alignment, no previous work has analysed the encoder-decoder attention behavior in speech translation (ST) nor used it to improve ST on a specific task. In this paper, we fill this gap by proposing an attention-based policy (EDAtt) for simultaneous ST (SimulST) that is motivated by an analysis of the existing attention relations between audio input and textual output. Its goal is to leverage the encoder-decoder attention scores to guide inference in real time. Results on en->{de, es} show that the EDAtt policy achieves overall better results compared to the SimulST state of the art, especially in terms of computational-aware latency.

African languages have far less in-language content available digitally, making it challenging for question answering systems to satisfy the information needs of users. Cross-lingual open-retrieval question answering (XOR QA) systems -- those that retrieve answer content from other languages while serving people in their native language -- offer a means of filling this gap. To this end, we create AfriQA, the first cross-lingual QA dataset with a focus on African languages. AfriQA includes 12,000+ XOR QA examples across 10 African languages. While previous datasets have focused primarily on languages where cross-lingual QA augments coverage from the target language, AfriQA focuses on languages where cross-lingual answer content is the only high-coverage source of answer content. Because of this, we argue that African languages are one of the most important and realistic use cases for XOR QA. Our experiments demonstrate the poor performance of automatic translation and multilingual retrieval methods. Overall, AfriQA proves challenging for state-of-the-art QA models. We hope that the dataset enables the development of more equitable QA technology.

Recent advancements in Natural Language Processing (NLP) has led to the proliferation of large pretrained language models. These models have been shown to yield good performance, using in-context learning, even on unseen tasks and languages. They have also been exposed as commercial APIs as a form of language-model-as-a-service, with great adoption. However, their performance on African languages is largely unknown. We present a preliminary analysis of commercial large language models on two tasks (machine translation and text classification) across eight African languages, spanning different language families and geographical areas. Our results suggest that commercial language models produce below-par performance on African languages. We also find that they perform better on text classification than machine translation. In general, our findings present a call-to-action to ensure African languages are well represented in commercial large language models, given their growing popularity.

For years the model performance in machine learning obeyed a power-law relationship with the model size. For the consideration of parameter efficiency, recent studies focus on increasing model depth rather than width to achieve better performance. In this paper, we study how model width affects the Transformer model through a parameter-efficient multi-path structure. To better fuse features extracted from different paths, we add three additional operations to each sublayer: a normalization at the end of each path, a cheap operation to produce more features, and a learnable weighted mechanism to fuse all features flexibly. Extensive experiments on 12 WMT machine translation tasks show that, with the same number of parameters, the shallower multi-path model can achieve similar or even better performance than the deeper model. It reveals that we should pay more attention to the multi-path structure, and there should be a balance between the model depth and width to train a better large-scale Transformer.

Text image machine translation (TIMT) has been widely used in various real-world applications, which translates source language texts in images into another target language sentence. Existing methods on TIMT are mainly divided into two categories: the recognition-then-translation pipeline model and the end-to-end model. However, how to transfer knowledge from the pipeline model into the end-to-end model remains an unsolved problem. In this paper, we propose a novel Multi-Teacher Knowledge Distillation (MTKD) method to effectively distillate knowledge into the end-to-end TIMT model from the pipeline model. Specifically, three teachers are utilized to improve the performance of the end-to-end TIMT model. The image encoder in the end-to-end TIMT model is optimized with the knowledge distillation guidance from the recognition teacher encoder, while the sequential encoder and decoder are improved by transferring knowledge from the translation sequential and decoder teacher models. Furthermore, both token and sentence-level knowledge distillations are incorporated to better boost the translation performance. Extensive experimental results show that our proposed MTKD effectively improves the text image translation performance and outperforms existing end-to-end and pipeline models with fewer parameters and less decoding time, illustrating that MTKD can take advantage of both pipeline and end-to-end models.