In natural language processing (NLP), the problem of aligning monolingual embedding spaces to induce a shared cross-lingual vector space has been shown not only to be useful in a variety of tasks such as bilingual lexicon induction (BLI) (Mikolov et al., 2013; Barone, 2016; Artetxe et al., 2017; Aboagye et al., 2022), machine translation (Artetxe et al., 2018b), cross-lingual information retrieval (Vulić & Moens, 2015), but it plays a crucial role in facilitating the cross-lingual transfer of language technologies from high resource languages to low resource languages. Cross-lingual word embeddings (CLWEs) represent words from two or more languages in a shared cross-lingual vector space in which words with similar meanings obtain similar vectors regardless of their language. There has been a flurry of work dominated by the so-called projection-based CLWE models (Mikolov et al., 2013; Artetxe et al., 2016, 2017, 2018a; Smith et al., 2017; Ruder et al., 2019), which aim to improve CLWE model performance significantly. Projection-based CLWE models learn a transfer function or mapper between two independently trained monolingual word vector spaces with limited or no cross-lingual supervision. Famous among projection-based CLWE models are the unsupervised projection-based CLWE models (Artetxe et al., 2017; Lample et al., 2018; Alvarez-Melis & Jaakkola, 2018;

We present a method for introducing a text encoder into pre-trained end-to-end speech translation systems. It enhances the ability of adapting one modality (i.e., source-language speech) to another (i.e., source-language text). Thus, the speech translation model can learn from both unlabeled and labeled data, especially when the source-language text data is abundant. Beyond this, we present a denoising method to build a robust text encoder that can deal with both normal and noisy text data. Our system sets new state-of-the-arts on the MuST-C En-De, En-Fr, and LibriSpeech En-Fr tasks.

This technical report briefly describes our JDExplore d-team's Vega v2 submission on the SuperGLUE leaderboard. SuperGLUE is more challenging than the widely used general language understanding evaluation (GLUE) benchmark, containing eight difficult language understanding tasks, including question answering, natural language inference, word sense disambiguation, coreference resolution, and reasoning. [Method] Instead of arbitrarily increasing the size of a pretrained language model (PLM), our aim is to 1) fully extract knowledge from the input pretraining data given a certain parameter budget, e.g., 6B, and 2) effectively transfer this knowledge to downstream tasks. To achieve goal 1), we propose self-evolution learning for PLMs to wisely predict the informative tokens that should be masked, and supervise the masked language modeling (MLM) process with rectified smooth labels. For goal 2), we leverage the prompt transfer technique to improve the low-resource tasks by transferring the knowledge from the foundation model and related downstream tasks to the target task. [Results] According to our submission record (Oct. 2022), with our optimized pretraining and fine-tuning strategies, our 6B Vega method achieved new state-of-the-art performance on 4/8 tasks, sitting atop the SuperGLUE leaderboard on Oct. 8, 2022, with an average score of 91.3.

We present the task of PreQuEL, Pre-(Quality-Estimation) Learning. A PreQuEL system predicts how well a given sentence will be translated, without recourse to the actual translation, thus eschewing unnecessary resource allocation when translation quality is bound to be low. PreQuEL can be defined relative to a given MT system (e.g., some industry service) or generally relative to the state-of-the-art. From a theoretical perspective, PreQuEL places the focus on the source text, tracing properties, possibly linguistic features, that make a sentence harder to machine translate. We develop a baseline model for the task and analyze its performance. We also develop a data augmentation method (from parallel corpora), that improves results substantially. We show that this augmentation method can improve the performance of the Quality-Estimation task as well. We investigate the properties of the input text that our model is sensitive to, by testing it on challenge sets and different languages. We conclude that it is aware of syntactic and semantic distinctions, and correlates and even over-emphasizes the importance of standard NLP features.

Link prediction (LP) has been recognized as an important task in graph learning with its broad practical applications. A typical application of LP is to retrieve the top scoring neighbors for a given source node, such as the friend recommendation. These services desire the high inference scalability to find the top scoring neighbors from many candidate nodes at low latencies. There are two popular decoders that the recent LP models mainly use to compute the edge scores from node embeddings: the HadamardMLP and Dot Product decoders. After theoretical and empirical analysis, we find that the HadamardMLP decoders are generally more effective for LP. However, HadamardMLP lacks the scalability for retrieving top scoring neighbors on large graphs, since to the best of our knowledge, there does not exist an algorithm to retrieve the top scoring neighbors for HadamardMLP decoders in sublinear complexity. To make HadamardMLP scalable, we propose the Flashlight algorithm to accelerate the top scoring neighbor retrievals for HadamardMLP: a sublinear algorithm that progressively applies approximate maximum inner product search (MIPS) techniques with adaptively adjusted query embeddings. Empirical results show that Flashlight improves the inference speed of LP by more than 100 times on the large OGBL-CITATION2 dataset without sacrificing effectiveness. Our work paves the way for large-scale LP applications with the effective HadamardMLP decoders by greatly accelerating their inference.

Unavailability of parallel corpora for training text style transfer (TST) models is a very challenging yet common scenario. Also, TST models implicitly need to preserve the content while transforming a source sentence into the target style. To tackle these problems, an intermediate representation is often constructed that is devoid of style while still preserving the meaning of the source sentence. In this work, we study the usefulness of Abstract Meaning Representation (AMR) graph as the intermediate style agnostic representation. We posit that semantic notations like AMR are a natural choice for an intermediate representation. Hence, we propose T-STAR: a model comprising of two components, text-to-AMR encoder and a AMR-to-text decoder. We propose several modeling improvements to enhance the style agnosticity of the generated AMR. To the best of our knowledge, T-STAR is the first work that uses AMR as an intermediate representation for TST. With thorough experimental evaluation we show T-STAR significantly outperforms state of the art techniques by achieving on an average 15.2% higher content preservation with negligible loss (3% approx.) in style accuracy. Through detailed human evaluation with 90,000 ratings, we also show that T-STAR has up to 50% lesser hallucinations compared to state of the art TST models.

This paper describes the submission of the RoyalFlush neural machine translation system for the WMT 2022 translation efficiency task. Unlike the commonly used autoregressive translation system, we adopted a two-stage translation paradigm called Hybrid Regression Translation (HRT) to combine the advantages of autoregressive and non-autoregressive translation. Specifically, HRT first autoregressively generates a discontinuous sequence (e.g., make a prediction every $k$ tokens, $k>1$) and then fills in all previously skipped tokens at once in a non-autoregressive manner. Thus, we can easily trade off the translation quality and speed by adjusting $k$. In addition, by integrating other modeling techniques (e.g., sequence-level knowledge distillation and deep-encoder-shallow-decoder layer allocation strategy) and a mass of engineering efforts, HRT improves 80\% inference speed and achieves equivalent translation performance with the same-capacity AT counterpart. Our fastest system reaches 6k+ words/second on the GPU latency setting, estimated to be about 3.1x faster than the last year's winner.

Interpretable entity representations (IERs) are sparse embeddings that are "human-readable" in that dimensions correspond to fine-grained entity types and values are predicted probabilities that a given entity is of the corresponding type. These methods perform well in zero-shot and low supervision settings. Compared to standard dense neural embeddings, such interpretable representations may permit analysis and debugging. However, while fine-tuning sparse, interpretable representations improves accuracy on downstream tasks, it destroys the semantics of the dimensions which were enforced in pre-training. Can we maintain the interpretable semantics afforded by IERs while improving predictive performance on downstream tasks? Toward this end, we propose Intermediate enTity-based Sparse Interpretable Representation Learning (ItsIRL). ItsIRL realizes improved performance over prior IERs on biomedical tasks, while maintaining "interpretability" generally and their ability to support model debugging specifically. The latter is enabled in part by the ability to perform "counterfactual" fine-grained entity type manipulation, which we explore in this work. Finally, we propose a method to construct entity type based class prototypes for revealing global semantic properties of classes learned by our model.

Artificial intelligence (AI) continuously wows or terrifies us, but there's no denying that AI will play an essential role in human development over the next decade. Machine translation, which has been around since the 1950s, will soon make extreme strides thanks to AI technologies. AI language translation is rooted in machine translation, which is a specialized technology that translates text without human assistance. While machine translation did come first, artificial intelligence translation and technology were developed side-by-side and aided their progress. That means that speech-to-text and the software that supports it have a symbiotic relationship.

Subword-level models have been the dominant paradigm in NLP. However, character-level models have the benefit of seeing each character individually, providing the model with more detailed information that ultimately could lead to better models. Recent works have shown character-level models to be competitive with subword models, but costly in terms of time and computation. Character-level models with a downsampling component alleviate this, but at the cost of quality, particularly for machine translation. This work analyzes the problems of previous downsampling methods and introduces a novel downsampling method which is informed by subwords. This new downsampling method not only outperforms existing downsampling methods, showing that downsampling characters can be done without sacrificing quality, but also leads to promising performance compared to subword models for translation.