In this work, we study how the generalization performance of a given direction changes with its sampling ratio in Multilingual Neural Machine Translation (MNMT). By training over 200 multilingual models with various model sizes, directions, and total numbers of tasks, we find that scalarization leads to a multitask trade-off front that deviates from the traditional Pareto front when there exists data imbalance in the training corpus. That is, the performance of certain translation directions does not improve with the increase of its weight in the multi-task optimization objective, which poses a great challenge to improve the overall performance of all directions. Based on our observations, we propose the Double Power Law to predict the unique performance trade-off front in MNMT, which is robust across various languages, data adequacy, and the number of tasks. Finally, we formulate the sample ratio selection problem in MNMT as an optimization problem based on the Double Power Law, which achieves better performance than temperature searching and gradient manipulation methods using up to half of the total training budget in our experiments.
Despite that going deep has proven successful in many neural architectures, the existing graph transformers are relatively shallow. In this work, we explore whether more layers are beneficial to graph transformers, and find that current graph transformers suffer from the bottleneck of improving performance by increasing depth. Our further analysis reveals the reason is that deep graph transformers are limited by the vanishing capacity of global attention, restricting the graph transformer from focusing on the critical substructure and obtaining expressive features. To this end, we propose a novel graph transformer model named DeepGraph that explicitly employs substructure tokens in the encoded representation, and applies local attention on related nodes to obtain substructure based attention encoding. Our model enhances the ability of the global attention to focus on substructures and promotes the expressiveness of the representations, addressing the limitation of self-attention as the graph transformer deepens. Experiments show that our method unblocks the depth limitation of graph transformers and results in state-of-the-art performance across various graph benchmarks with deeper models.
Context-aware neural machine translation aims to use the document-level context to improve translation quality. However, not all words in the context are helpful. The irrelevant or trivial words may bring some noise and distract the model from learning the relationship between the current sentence and the auxiliary context. To mitigate this problem, we propose a novel end-to-end encoder-decoder model with a layer-wise selection mechanism to sift and refine the long document context. To verify the effectiveness of our method, extensive experiments and extra quantitative analysis are conducted on four document-level machine translation benchmarks. The experimental results demonstrate that our model significantly outperforms previous models on all datasets via the soft selection mechanism.
Pre-trained models have achieved remarkable success in natural language processing (NLP). However, existing pre-training methods underutilize the benefits of language understanding for generation. Inspired by the idea of Generative Adversarial Networks (GANs), we propose a GAN-style model for encoder-decoder pre-training by introducing an auxiliary discriminator, unifying the ability of language understanding and generation in a single model. Our model, named as GanLM, is trained with two pre-training objectives: replaced token detection and replaced token denoising. Specifically, given masked source sentences, the generator outputs the target distribution and the discriminator predicts whether the target sampled tokens from distribution are incorrect. The target sentence is replaced with misclassified tokens to construct noisy previous context, which is used to generate the gold sentence. In general, both tasks improve the ability of language understanding and generation by selectively using the denoising data. Extensive experiments in language generation benchmarks show that GanLM with the powerful language understanding capability outperforms various strong pre-trained language models (PLMs) and achieves state-of-the-art performance.
Despite the current success of multilingual pre-training, most prior works focus on leveraging monolingual data or bilingual parallel data and overlooked the value of trilingual parallel data. This paper presents \textbf{Tri}angular Document-level \textbf{P}re-training (\textbf{TRIP}), which is the first in the field to extend the conventional monolingual and bilingual pre-training to a trilingual setting by (i) \textbf{Grafting} the same documents in two languages into one mixed document, and (ii) predicting the remaining one language as the reference translation. Our experiments on document-level MT and cross-lingual abstractive summarization show that TRIP brings by up to 3.65 d-BLEU points and 6.2 ROUGE-L points on three multilingual document-level machine translation benchmarks and one cross-lingual abstractive summarization benchmark, including multiple strong state-of-the-art (SOTA) scores. In-depth analysis indicates that TRIP improves document-level machine translation and captures better document contexts in at least three characteristics: (i) tense consistency, (ii) noun consistency and (iii) conjunction presence.
Machine translation (MT) has almost achieved human parity at sentence-level translation. In response, the MT community has, in part, shifted its focus to document-level translation. However, the development of document-level MT systems is hampered by the lack of parallel document corpora. This paper describes BWB, a large parallel corpus first introduced in Jiang et al. (2022), along with an annotated test set. The BWB corpus consists of Chinese novels translated by experts into English, and the annotated test set is designed to probe the ability of machine translation systems to model various discourse phenomena. Our resource is freely available, and we hope it will serve as a guide and inspiration for more work in document-level machine translation.
Named entity recognition (NER) suffers from the scarcity of annotated training data, especially for low-resource languages without labeled data. Cross-lingual NER has been proposed to alleviate this issue by transferring knowledge from high-resource languages to low-resource languages via aligned cross-lingual representations or machine translation results. However, the performance of cross-lingual NER methods is severely affected by the unsatisfactory quality of translation or label projection. To address these problems, we propose a Cross-lingual Entity Projection framework (CROP) to enable zero-shot cross-lingual NER with the help of a multilingual labeled sequence translation model. Specifically, the target sequence is first translated into the source language and then tagged by a source NER model. We further adopt a labeled sequence translation model to project the tagged sequence back to the target language and label the target raw sentence. Ultimately, the whole pipeline is integrated into an end-to-end model by the way of self-training. Experimental results on two benchmarks demonstrate that our method substantially outperforms the previous strong baseline by a large margin of +3~7 F1 scores and achieves state-of-the-art performance.
Multilingual machine translation has been proven an effective strategy to support translation between multiple languages with a single model. However, most studies focus on multilingual sentence translation without considering generating long documents across different languages, which requires an understanding of multilingual context dependency and is typically harder. In this paper, we first spot that naively incorporating auxiliary multilingual data either auxiliary-target or source-auxiliary brings no improvement to the source-target language pair in our interest. Motivated by this observation, we propose a novel framework called Multilingual Transitivity (MTrans) to find an implicit optimal route via source-auxiliary-target within the multilingual model. To encourage MTrans, we propose a novel method called Triplet Parallel Data (TPD), which uses parallel triplets that contain (source-auxiliary, auxiliary-target, and source-target) for training. The auxiliary language then serves as a pivot and automatically facilitates the implicit information transition flow which is easier to translate. We further propose a novel framework called Bidirectional Multilingual Agreement (Bi-MAgree) that encourages the bidirectional agreement between different languages. To encourage Bi-MAgree, we propose a novel method called Multilingual Kullback-Leibler Divergence (MKL) that forces the output distribution of the inputs with the same meaning but in different languages to be consistent with each other. The experimental results indicate that our methods bring consistent improvements over strong baselines on three document translation tasks: IWSLT2015 Zh-En, De-En, and Vi-En. Our analysis validates the usefulness and existence of MTrans and Bi-MAgree, and our frameworks and methods are effective on synthetic auxiliary data.
Multilingual machine translation has been proven an effective strategy to support translation between multiple languages with a single model. However, most studies focus on multilingual sentence translation without considering generating long documents across different languages, which requires an understanding of multilingual context dependency and is typically harder. In this paper, we first spot that naively incorporating auxiliary multilingual data either auxiliary-target or source-auxiliary brings no improvement to the source-target language pair in our interest. Motivated by this observation, we propose a novel framework called Multilingual Transitivity (MTrans) to find an implicit optimal route via source-auxiliary-target within the multilingual model. To encourage MTrans, we propose a novel method called Triplet Parallel Data (TPD), which uses parallel triplets that contain (source-auxiliary, auxiliary-target, and source-target) for training. The auxiliary language then serves as a pivot and automatically facilitates the implicit information transition flow which is easier to translate. We further propose a novel framework called Bidirectional Multilingual Agreement (Bi-MAgree) that encourages the bidirectional agreement between different languages. To encourage Bi-MAgree, we propose a novel method called Multilingual Kullback-Leibler Divergence (MKL) that forces the output distribution of the inputs with the same meaning but in different languages to be consistent with each other. The experimental results indicate that our methods bring consistent improvements over strong baselines on three document translation tasks: IWSLT2015 Zh-En, De-En, and Vi-En. Our analysis validates the usefulness and existence of MTrans and Bi-MAgree, and our frameworks and methods are effective on synthetic auxiliary data.
Transformer structure, stacked by a sequence of encoder and decoder network layers, achieves significant development in neural machine translation. However, vanilla Transformer mainly exploits the top-layer representation, assuming the lower layers provide trivial or redundant information and thus ignoring the bottom-layer feature that is potentially valuable. In this work, we propose the Group-Transformer model (GTrans) that flexibly divides multi-layer representations of both encoder and decoder into different groups and then fuses these group features to generate target words. To corroborate the effectiveness of the proposed method, extensive experiments and analytic experiments are conducted on three bilingual translation benchmarks and two multilingual translation tasks, including the IWLST-14, IWLST-17, LDC, WMT-14 and OPUS-100 benchmark. Experimental and analytical results demonstrate that our model outperforms its Transformer counterparts by a consistent gain. Furthermore, it can be successfully scaled up to 60 encoder layers and 36 decoder layers.