Machine Learning (ML) interatomic models and potentials have been widely employed in simulations of materials. Long-range interactions often dominate in some ionic systems whose dynamics behavior is significantly influenced. However, the long-range effect such as Coulomb and Van der Wales potential is not considered in most ML interatomic potentials. To address this issue, we put forward a method that can take long-range effects into account for most ML local interatomic models with the reciprocal space neural network. The structure information in real space is firstly transformed into reciprocal space and then encoded into a reciprocal space potential or a global descriptor with full atomic interactions. The reciprocal space potential and descriptor keep full invariance of Euclidean symmetry and choice of the cell. Benefiting from the reciprocal-space information, ML interatomic models can be extended to describe the long-range potential including not only Coulomb but any other long-range interaction. A model NaCl system considering Coulomb interaction and the GaxNy system with defects are applied to illustrate the advantage of our approach. At the same time, our approach helps to improve the prediction accuracy of some global properties such as the band gap where the full atomic interaction beyond local atomic environments plays a very important role. In summary, our work has expanded the ability of current ML interatomic models and potentials when dealing with the long-range effect, hence paving a new way for accurate prediction of global properties and large-scale dynamic simulations of systems with defects.

This work presents Time-reversal Equivariant Neural Network (TENN) framework. With TENN, the time-reversal symmetry is considered in the equivariant neural network (ENN), which generalizes the ENN to consider physical quantities related to time-reversal symmetry such as spin and velocity of atoms. TENN-e3, as the time-reversal-extension of E(3) equivariant neural network, is developed to keep the Time-reversal E(3) equivariant with consideration of whether to include the spin-orbit effect for both collinear and non-collinear magnetic moments situations for magnetic material. TENN-e3 can construct spin neural network potential and the Hamiltonian of magnetic material from ab-initio calculations. Time-reversal-E(3)-equivariant convolutions for interactions of spinor and geometric tensors are employed in TENN-e3. Compared to the popular ENN, TENN-e3 can describe the complex spin-lattice coupling with high accuracy and keep time-reversal symmetry which is not preserved in the existing E(3)-equivariant model. Also, the Hamiltonian of magnetic material with time-reversal symmetry can be built with TENN-e3. TENN paves a new way to spin-lattice dynamics simulations over long-time scales and electronic structure calculations of large-scale magnetic materials.

We present SpeechMatrix, a large-scale multilingual corpus of speech-to-speech translations mined from real speech of European Parliament recordings. It contains speech alignments in 136 language pairs with a total of 418 thousand hours of speech. To evaluate the quality of this parallel speech, we train bilingual speech-to-speech translation models on mined data only and establish extensive baseline results on EuroParl-ST, VoxPopuli and FLEURS test sets. Enabled by the multilinguality of SpeechMatrix, we also explore multilingual speech-to-speech translation, a topic which was addressed by few other works. We also demonstrate that model pre-training and sparse scaling using Mixture-of-Experts bring large gains to translation performance. The mined data and models are freely available.

We present a new approach to perform zero-shot cross-modal transfer between speech and text for translation tasks. Multilingual speech and text are encoded in a joint fixed-size representation space. Then, we compare different approaches to decode these multimodal and multilingual fixed-size representations, enabling zero-shot translation between languages and modalities. All our models are trained without the need of cross-modal labeled translation data. Despite a fixed-size representation, we achieve very competitive results on several text and speech translation tasks. In particular, we significantly improve the state-of-the-art for zero-shot speech translation on Must-C. Incorporating a speech decoder in our framework, we introduce the first results for zero-shot direct speech-to-speech and text-to-speech translation.

惯用表达式（IES）在自然语言中起重要作用。在本文中，我们研究了惯用句子解释（ISP）的任务，旨在通过用IE用文字解释来解释一个句子。缺乏与惯用语文平行句子的大型语料库是这项任务的主要挑战，我们考虑了两个单独的解决方案。首先，我们向ISP提出了一个无人监督的方法，它利用IE的上下文信息和定义，不需要并行句子训练集。其次，我们提出了一种弱监督的方法，使用后翻来的方法与IE共同执行释义和生成句子，以扩大小规模并行句子训练数据集。该研究的其他重要衍生物包括一种模型，该模型将句子中的文字短语替换为一种与IE生成惯用表达式和具有惯用/文字句对的大规模并行数据集。拟议的解决方案与竞争性基线相比的有效性在Bleu超过5.16点的相对增益中观察到超过8.75点，在使用自动和手动的并行数据集上经验上验证生成的句子时，Sari超过19.57点评估。我们展示了ISP作为EN-DE机器翻译中的预处理步骤的实用实用性。

我们介绍了一种无线文字语音转换（S2ST）系统，可以将来自一种语言的语音转换为另一种语言，并且可以在不需要任何文本数据的情况下构建。与文献中的现有工作不同，我们解决了模拟多扬声器目标语音的挑战，并用现实世界的S2ST数据训练系统。我们方法的关键是一种自我监督的单位语音标准化技术，该标准化技术将预先训练的语音编码器具有来自多个扬声器的配对声音，以及单个参考扬声器，以减少由于复印件引起的变化，同时保留词汇内容。只有10分钟的语音标准化的配对数据，我们在培训\ vp〜s2st数据集上的S2ST模型时获得平均3.2 BLEU增益，而不是在未标准化的语音目标上培训的基线。我们还将自动开采的S2ST数据纳入并显示额外的2.0 BLEU增益。据我们所知，我们是第一个建立无线的S2ST技术，可以用真实世界的数据培训，并为多种语言配对工作。

我们提出了直接同时的语音转换（SIMUL-S2ST）模型，此外，翻译的产生与中间文本表示无关。我们的方法利用了最近与离散单位直接语音转换的最新进展，其中从模型中预测了一系列离散表示，而不是连续频谱图特征，而不是以无监督的方式学习，并直接传递给语音的声码器综合在一起。我们还介绍了变分单调的多口语注意力（V-MMA），以处理语音同声翻译中效率低效的政策学习的挑战。然后，同时策略在源语音特征和目标离散单元上运行。我们开展实证研究，比较级联和直接方法对Fisher西班牙语 - 英语和必需的英语西班牙语数据集。直接同步模型显示通过在翻译质量和延迟之间实现更好的权衡来优于级联模型。

语音到语音翻译（S2ST）将输入语音转换为另一种语言。实时交付S2ST的挑战是翻译和语音合成模块之间的累积延迟。尽管最近增量的文本到语音（ITTS）模型已显示出巨大的质量改进，但它们通常需要其他未来的文本输入才能达到最佳性能。在这项工作中，我们通过调整上游语音翻译器来为语音合成器生成高质量的伪lookahead来最大程度地减少ITT的最初等待时间。缓解初始延迟后，我们证明了合成语音的持续时间在延迟中也起着至关重要的作用。我们将其形式化为延迟度量，然后提出一种简单而有效的持续时间缩放方法，以减少延迟。我们的方法始终将延迟减少0.2-0.5秒，而无需牺牲语音翻译质量。

多语种模型是参数效率，特别是通过利用Crosslingual Transcer来改善低资源语言。尽管最近有巨大的多语言翻译预先推出了越来越多的模型和数据，但如何有效地培养多语言模型并未得到很好的理解。在本文中，我们表明，多语言训练中的常见情况，语言之间的数据不平衡，高资源和低资源语言之间的优化张力，其中发现的多语言解决方案通常是低资源的次优。我们展示了普通培训方法，upsamples低资源无法鲁布利地优化人口损失，其中风险耗材或过度为低资源的风险。绘制最近关于损失景观几何学的发现及其对泛化的影响，提出了一个原则性的优化算法，曲率意识的任务缩放（CAT），其自适应地从不同任务中重新加强了对低曲率的多语言训练的元的梯度。邻居对所有语言均匀低损失。我们在共同基准（TED，WMT和OPUS-100）上进行了实验，具有不同程度的数据不平衡。猫有效地改善了多语言优化，结果表明，在低资源（$ + 0.8 $至+ 2.2 $ BLEU）上展示了一致的收益，而不会伤害高资源。此外，猫对过度分数计量和大量批量训练具有强大的稳健性，这使得这是一种充满希望的大量多语言模型，真正提高低资源语言。

Projection operations are a typical computation bottleneck in online learning. In this paper, we enable projection-free online learning within the framework of Online Convex Optimization with Memory (OCO-M) -- OCO-M captures how the history of decisions affects the current outcome by allowing the online learning loss functions to depend on both current and past decisions. Particularly, we introduce the first projection-free meta-base learning algorithm with memory that minimizes dynamic regret, i.e., that minimizes the suboptimality against any sequence of time-varying decisions. We are motivated by artificial intelligence applications where autonomous agents need to adapt to time-varying environments in real-time, accounting for how past decisions affect the present. Examples of such applications are: online control of dynamical systems; statistical arbitrage; and time series prediction. The algorithm builds on the Online Frank-Wolfe (OFW) and Hedge algorithms. We demonstrate how our algorithm can be applied to the online control of linear time-varying systems in the presence of unpredictable process noise. To this end, we develop the first controller with memory and bounded dynamic regret against any optimal time-varying linear feedback control policy. We validate our algorithm in simulated scenarios of online control of linear time-invariant systems.