Research on text-to-image generation (TTI) still predominantly focuses on the English language due to the lack of annotated imagecaption data in other languages; in the long run, this might widen inequitable access to TTI technology.In this work, we thus investigate multilingual TTI (termed mTTI) and the current potential of neural machine translation (NMT) to bootstrap mTTI systems.We provide two key contributions.1) Relying on a multilingual multi-modal encoder, we provide a systematic empirical study of standard methods used in cross-lingual NLP when applied to mTTI: TRANSLATE TRAIN, TRANS-LATE TEST, and ZERO-SHOT TRANSFER.2) We propose Ensemble Adapter (ENSAD), a novel parameter-efficient approach that learns to weigh and consolidate the multilingual text knowledge within the mTTI framework, mitigating the language gap and thus improving mTTI performance.Our evaluations on standard mTTI datasets COCO-CN, Multi30K Task2, and LAION-5B demonstrate the potential of translation-enhanced mTTI systems and also validate the benefits of the proposed EN-SAD which derives consistent gains across all datasets.Further investigations on model variants, ablation studies, and qualitative analyses provide additional insights on the inner workings of the proposed mTTI approaches.

Texts from the Internet serve as important data sources for financial market modeling. Early statistical approaches rely on manually defined features to capture lexical, sentiment and event information, which suffers from feature sparsity. Recent work has considered learning dense representations for news titles and abstracts. Compared to news titles, full documents can contain more potentially helpful information, but also noise compared to events and sentences, which has been less investigated in previous work. To fill this gap, we propose a novel target-specific abstract-guided news document representation model. The model uses a target-sensitive representation of the news abstract to weigh sentences in the news content, so as to select and combine the most informative sentences for market modeling. Results show that document representations can give better performance for estimating cumulative abnormal returns of companies when compared to titles and abstracts. Our model is especially effective when it used to combine information from multiple document sources compared to the sentence-level baselines.

There has been a recent line of work automatically learning scripts from unstructured texts, by modeling narrative event chains.While the dominant approach group events using event pair relations, L-STMs have been used to encode full chains of narrative events.The latter has the advantage of learning long-range temporal orders 1 , yet the former is more adaptive to partial orders.We propose a neural model that leverages the advantages of both methods, by using LSTM hidden states as features for event pair modelling.A dynamic memory network is utilized to automatically induce weights on existing events for inferring a subsequent event.Standard evaluation shows that our method significantly outperforms both methods above, giving the best results reported so far.

Measuring the information content of news text is useful for decision makers in their investments since news information can influence the intrinsic values of companies. We propose a model to automatically measure the information content given news text, trained using news and corresponding cumulative abnormal returns of listed companies. Existing methods in finance literature exploit sentiment signal features, which are limited by not considering factors such as events. We address this issue by leveraging deep neural models to extract rich semantic features from news text. In particular, a novel tree-structured LSTM is used to find target-specific representations of news text given syntax structures. Empirical results show that the neural models can outperform sentiment-based models, demonstrating the effectiveness of recent NLP technology advances for computational finance.

Many proteomic investigations summarize the quantitative information across multiple spectral features into protein-level conclusions. Data-independent spectral acquisition (DIA) now generates a lot of interest, as it allows us to quantify many spectral features in a single run. However, the disadvantage of DIA experiments as compared, e.g., to Selected Reaction Monitoring (SRM) is that the features are subject to interferences and noise. We argue that between-run variation provides an additional insight for distinguishing good-quality and noisy DIA features. To appropriately use the quantitative between-run variation, it is important to account for the properties experimental design, and distinguish random artifacts from the biological changes. We have previously proposed a method (Chang et al., ASMS 2013) that accounts for the experimental design to eliminate features with low information content. In this project we furthermore emphasized that conducting regularization helps us avoid exploring every subset of features exhaustively, and allows us to conduct hypothesis tests later on so that we would be able to control the false discovery rate of the feature selection process. We evaluated our proposed approach by using three datasets that have some notion of ground truth: an extensive simulation study, a controlled mixture where proteins were spiked into a complex background in known concentrations, and a study of 232 plasma samples, where 18 proteins were quantified in both SWAH and SRM mode in presence of heavy labeled reference peptides. We worked on [ 1 ] protein-level estimates of fold changes between conditions, [ 2 ] sensitivity and specificity of detecting changes in protein abundance, and [ 3 ] accuracy of relative quantification of protein abundance in individual biological samples. A family of linear mixed models similar to that in MSstats http://www.msstats.org were fit to all the datasets. Then we conducted the regularization and hypothesis test to control the selection false discovery rate. The results demonstrated that our proposed feature selection approach enhanced sensitivity and specificity of the conclusions, was robust to the amount of noisy fragments, and increased the correlation of subject quantification between SRM and DIA workflows. Importantly, the performance exceeded that of the frequently used 'top 3' approach, which consists of using three spectral features with the highest average intensity between runs. Furthermore, we showed that our proposed approach outperforms using correlation to select the information features.

UNLABELLED: MSstats is an R package for statistical relative quantification of proteins and peptides in mass spectrometry-based proteomics. Version 2.0 of MSstats supports label-free and label-based experimental workflows and data-dependent, targeted and data-independent spectral acquisition. It takes as input identified and quantified spectral peaks, and outputs a list of differentially abundant peptides or proteins, or summaries of peptide or protein relative abundance. MSstats relies on a flexible family of linear mixed models. AVAILABILITY AND IMPLEMENTATION: The code, the documentation and example datasets are available open-source at www.msstats.org under the Artistic-2.0 license. The package can be downloaded from www.msstats.org or from Bioconductor www.bioconductor.org and used in an R command line workflow. The package can also be accessed as an external tool in Skyline (Broudy et al., 2014) and used via graphical user interface.

Linguistic steganography is concerned with hiding information in natural language text. One of the major transformations used in linguistic steganography is synonym substitution. However, few existing studies have studied the practical application of this approach. In this article we propose two improvements to the use of synonym substitution for encoding hidden bits of information. First, we use the Google n-gram corpus for checking the applicability of a synonym in context, and we evaluate this method using data from the SemEval lexical substitution task and human annotated data. Second, we address the problem that arises from words with more than one sense, which creates a potential ambiguity in terms of which bits are represented by a particular word. We develop a novel method in which words are the vertices in a graph, synonyms are linked by edges, and the bits assigned to a word are determined by a vertex coding algorithm. This method ensures that each word represents a unique sequence of bits, without cutting out large numbers of synonyms, and thus maintains a reasonable embedding capacity.

We invented a progressive image transmission scheme that is used for transmitting Chinese calligraphy. The scheme employs the property of simple colors of calligraphy images to design a method of transmitting images phase by phase. Overall, our scheme can achieve the following two goals. One is compressing the image data to reduce the transmission time while the other is gaining less response time by using progressive image transmission. Furthermore, the recovered image still maintains the colors of the seals with high image quality.