Bilingual lexicon induction (BLI) with limited bilingual supervision is a crucial yet challenging task in multilingual NLP. Current state-of-the-art BLI methods rely on the induction of cross-lingual word embeddings (CLWEs) to capture cross-lingual word similarities; such CLWEs are obtained 1) via traditional static models (e.g., VecMap), or 2) by extracting type-level CLWEs from multilingual pretrained language models (mPLMs), or 3) through combining the former two options. In this work, we propose a novel semi-supervised post-hoc reranking method termed BLICEr (BLI with Cross-Encoder Reranking), applicable to any precalculated CLWE space, which improves their BLI capability. The key idea is to 'extract' cross-lingual lexical knowledge from mPLMs, and then combine it with the original CLWEs. This crucial step is done via 1) creating a word similarity dataset, comprising positive word pairs (i.e., true translations) and hard negative pairs induced from the original CLWE space, and then 2) fine-tuning an mPLM (e.g., mBERT or XLM-R) in a cross-encoder manner to predict the similarity scores. At inference, we 3) combine the similarity score from the original CLWE space with the score from the BLI-tuned cross-encoder. BLICEr establishes new state-of-the-art results on two standard BLI benchmarks spanning a wide spectrum of diverse languages: it substantially outperforms a series of strong baselines across the board. We also validate the robustness of BLICEr with different CLWEs.

Scene marketing that well demonstrates user interests within a certain scenario has proved effective for offline shopping. To conduct scene marketing for e-commerce platforms, this work presents a novel product form, scene-based topic channel which typically consists of a list of diverse products belonging to the same usage scenario and a topic title that describes the scenario with marketing words. As manual construction of channels is time-consuming due to billions of products as well as dynamic and diverse customers' interests, it is necessary to leverage AI techniques to automatically construct channels for certain usage scenarios and even discover novel topics. To be specific, we first frame the channel construction task as a two-step problem, i.e., scene-based topic generation and product clustering, and propose an E-commerce Scene-based Topic Channel construction system (i.e., ESTC) to achieve automated production, consisting of scene-based topic generation model for the e-commerce domain, product clustering on the basis of topic similarity, as well as quality control based on automatic model filtering and human screening. Extensive offline experiments and online A/B test validates the effectiveness of such a novel product form as well as the proposed system. In addition, we also introduce the experience of deploying the proposed system on a real-world e-commerce recommendation platform.

"Prediction" is a broad and generic term that describes any process for obtaining guesses of unseen variables based on any available information, as well as each of these guesses. On the other hand, "forecasting" is a more specific term that describes any process for issuing predictions for future variables based on information (which most commonly takes the form of time series) about the present and the past, with these particular predictions being broadly called "forecasts". Forecasting is a key theme and topic for this study. Therefore, in what follows, the general focus will be on it and not on prediction in general, although many of the statements and methods that will be referring to it are equally relevant and applicable to other prediction types. The origins of forecasting trace back to the early humans and their pronounced need for certainty in the practical endeavour of supporting their various everyday life decisions (Petropoulos et al. 2022). Thus, forecasting has met until today and still meets numerous implementations, formal and informal. Independently of their exact categorization and features, the formal implementations of forecasting rely, in principal, on concepts, theory and practice that originate from or can be attributed to the predictive branch of statistical modelling, although forecasting is also considered as an entire field on its own because of the major role that the temporal dependence plays in the formulation of its methods. The predictive branch of statistical modelling exhibits profound and fundamental differences with respect to the descriptive and explanatory ones, as it is thoroughly explained in Shmueli (2010).

Large language models produce human-like text that drives a growing number of applications. However, recent literature and, increasingly, real world observations, have demonstrated that these models can generate language that is toxic, biased, untruthful or otherwise harmful. Though work to evaluate language model harms is under way, translating foresight about which harms may arise into rigorous benchmarks is not straightforward. To facilitate this translation, we outline six ways of characterizing harmful text which merit explicit consideration when designing new benchmarks. We then use these characteristics as a lens to identify trends and gaps in existing benchmarks. Finally, we apply them in a case study of the Perspective API, a toxicity classifier that is widely used in harm benchmarks. Our characteristics provide one piece of the bridge that translates between foresight and effective evaluation.

We study the text generation task under the approach of pre-trained language models (PLMs). Typically, an auto-regressive (AR) method is adopted for generating texts in a token-by-token manner. Despite many advantages of AR generation, it usually suffers from inefficient inference. Therefore, non-autoregressive (NAR) models are proposed to generate all target tokens simultaneously. However, NAR models usually generate texts of lower quality due to the absence of token dependency in the output text. In this paper, we propose ELMER: an efficient and effective PLM for NAR text generation to explicitly model the token dependency during NAR generation. By leveraging the early exit technique, ELMER enables the token generations at different layers, according to their prediction confidence (a more confident token will exit at a lower layer). Besides, we propose a novel pre-training objective, Layer Permutation Language Modeling, to pre-train ELMER by permuting the exit layer for each token in sequences. Experiments on three text generation tasks show that ELMER significantly outperforms NAR models and further narrows the performance gap with AR PLMs (\eg ELMER (29.92) vs BART (30.61) ROUGE-L in XSUM) while achieving over 10 times inference speedup.

A key bottleneck in building automatic extraction models for visually rich documents like invoices is the cost of acquiring the several thousand high-quality labeled documents that are needed to train a model with acceptable accuracy. We propose Selective Labeling to simplify the labeling task to provide "yes/no" labels for candidate extractions predicted by a model trained on partially labeled documents. We combine this with a custom active learning strategy to find the predictions that the model is most uncertain about. We show through experiments on document types drawn from 3 different domains that selective labeling can reduce the cost of acquiring labeled data by $10\times$ with a negligible loss in accuracy.

This paper reviews gradient-based techniques to solve bilevel optimization problems. Bilevel optimization is a general way to frame the learning of systems that are implicitly defined through a quantity that they minimize. This characterization can be applied to neural networks, optimizers, algorithmic solvers and even physical systems, and allows for greater modeling flexibility compared to an explicit definition of such systems. Here we focus on gradient-based approaches that solve such problems. We distinguish them in two categories: those rooted in implicit differentiation, and those that leverage the equilibrium propagation theorem. We present the mathematical foundations that are behind such methods, introduce the gradient-estimation algorithms in detail and compare the competitive advantages of the different approaches.

Although supervised deep learning has revolutionized speech and audio processing, it has necessitated the building of specialist models for individual tasks and application scenarios. It is likewise difficult to apply this to dialects and languages for which only limited labeled data is available. Self-supervised representation learning methods promise a single universal model that would benefit a wide variety of tasks and domains. Such methods have shown success in natural language processing and computer vision domains, achieving new levels of performance while reducing the number of labels required for many downstream scenarios. Speech representation learning is experiencing similar progress in three main categories: generative, contrastive, and predictive methods. Other approaches rely on multi-modal data for pre-training, mixing text or visual data streams with speech. Although self-supervised speech representation is still a nascent research area, it is closely related to acoustic word embedding and learning with zero lexical resources, both of which have seen active research for many years. This review presents approaches for self-supervised speech representation learning and their connection to other research areas. Since many current methods focus solely on automatic speech recognition as a downstream task, we review recent efforts on benchmarking learned representations to extend the application beyond speech recognition.

Deep Learning and big data have shown tremendous success in bioinformatics and computational biology in recent years; artificial intelligence methods have also significantly contributed in the task of protein function classification. This review paper analyzes the recent developments in approaches for the task of predicting protein function using deep learning. We explain the importance of determining the protein function and why automating the following task is crucial. Then, after reviewing the widely used deep learning techniques for this task, we continue our review and highlight the emergence of the modern State of The Art (SOTA) deep learning models which have achieved groundbreaking results in the field of computer vision, natural language processing and multi-modal learning in the last few years. We hope that this review will provide a broad view of the current role and advances of deep learning in biological sciences, especially in predicting protein function tasks and encourage new researchers to contribute to this area.

The facility location problems (FLPs) are a typical class of NP-hard combinatorial optimization problems, which are widely seen in the supply chain and logistics. Many mathematical and heuristic algorithms have been developed for optimizing the FLP. In addition to the transportation cost, there are usually multiple conflicting objectives in realistic applications. It is therefore desirable to design algorithms that find a set of Pareto solutions efficiently without enormous search cost. In this paper, we consider the multi-objective facility location problem (MO-FLP) that simultaneously minimizes the overall cost and maximizes the system reliability. We develop a learning-based approach to predicting the distribution probability of the entire Pareto set for a given problem. To this end, the MO-FLP is modeled as a bipartite graph optimization problem and two graph neural networks are constructed to learn the implicit graph representation on nodes and edges. The network outputs are then converted into the probability distribution of the Pareto set, from which a set of non-dominated solutions can be sampled non-autoregressively. Experimental results on MO-FLP instances of different scales show that the proposed approach achieves a comparable performance to a widely used multi-objective evolutionary algorithm in terms of the solution quality while significantly reducing the computational cost for search.